Water Production and Wastewater Treatment [1 ed.] 9781617610493, 9781617285035

Citation preview

Copyright © 2010. Nova Science Publishers, Incorporated. All rights reserved. Water Production and Wastewater Treatment, Nova Science Publishers, Incorporated, 2010. ProQuest Ebook Central,

Copyright © 2010. Nova Science Publishers, Incorporated. All rights reserved. Water Production and Wastewater Treatment, Nova Science Publishers, Incorporated, 2010. ProQuest Ebook Central,

WATER RESOURCE PLANNING, DEVELOPMENT AND MANAGEMENT

Copyright © 2010. Nova Science Publishers, Incorporated. All rights reserved.

WATER PRODUCTION AND WASTEWATER TREATMENT

No part of this digital document may be reproduced, stored in a retrieval system or transmitted in any form or by any means. The publisher has taken reasonable care in the preparation of this digital document, but makes no expressed or implied warranty of any kind and assumes no responsibility for any errors or omissions. No liability is assumed for incidental or consequential damages in connection with or arising out of information contained herein. This digital document is sold with the clear understanding that the publisher is not engaged in rendering legal, medical or any other professional services. Water Production and Wastewater Treatment, Nova Science Publishers, Incorporated, 2010. ProQuest Ebook Central,

WATER RESOURCE PLANNING, DEVELOPMENT AND MANAGEMENT Additional books in this series can be found on Nova’s website under the Series tab.

Additional E-books in this series can be found on Nova’s website under the E-books tab.

WASTE AND WASTE MANAGEMENT Additional books in this series can be found on Nova’s website under the Series tab.

Copyright © 2010. Nova Science Publishers, Incorporated. All rights reserved.

Additional E-books in this series can be found on Nova’s website under the E-books tab.

Water Production and Wastewater Treatment, Nova Science Publishers, Incorporated, 2010. ProQuest Ebook Central,

WATER RESOURCE PLANNING, DEVELOPMENT AND MANAGEMENT

WATER PRODUCTION AND WASTEWATER TREATMENT

B. ANTIZAR-LADISLAO AND Copyright © 2010. Nova Science Publishers, Incorporated. All rights reserved.

R. SHEIKHOLESLAMI EDITORS

Nova Science Publishers, Inc. New York

Water Production and Wastewater Treatment, Nova Science Publishers, Incorporated, 2010. ProQuest Ebook Central,

Copyright © 2011 by Nova Science Publishers, Inc. All rights reserved. No part of this book may be reproduced, stored in a retrieval system or transmitted in any form or by any means: electronic, electrostatic, magnetic, tape, mechanical photocopying, recording or otherwise without the written permission of the Publisher. For permission to use material from this book please contact us: Telephone 631-231-7269; Fax 631-231-8175 Web Site: http://www.novapublishers.com NOTICE TO THE READER The Publisher has taken reasonable care in the preparation of this book, but makes no expressed or implied warranty of any kind and assumes no responsibility for any errors or omissions. No liability is assumed for incidental or consequential damages in connection with or arising out of information contained in this book. The Publisher shall not be liable for any special, consequential, or exemplary damages resulting, in whole or in part, from the readers’ use of, or reliance upon, this material. Any parts of this book based on government reports are so indicated and copyright is claimed for those parts to the extent applicable to compilations of such works.

Copyright © 2010. Nova Science Publishers, Incorporated. All rights reserved.

Independent verification should be sought for any data, advice or recommendations contained in this book. In addition, no responsibility is assumed by the publisher for any injury and/or damage to persons or property arising from any methods, products, instructions, ideas or otherwise contained in this publication. This publication is designed to provide accurate and authoritative information with regard to the subject matter covered herein. It is sold with the clear understanding that the Publisher is not engaged in rendering legal or any other professional services. If legal or any other expert assistance is required, the services of a competent person should be sought. FROM A DECLARATION OF PARTICIPANTS JOINTLY ADOPTED BY A COMMITTEE OF THE AMERICAN BAR ASSOCIATION AND A COMMITTEE OF PUBLISHERS. Additional color graphics may be available in the e-book version of this book. LIBRARY OF CONGRESS CATALOGING-IN-PUBLICATION DATA Water production and wastewater treatment / editors, B. Antizar-Ladislao and R. Sheikholeslami. p. cm. Includes bibliographical references and index. ISBN:  (eBook) 1. Water--Purification. 2. Water-supply. I. Antizar-Ladislao, B. II. Sheikholeslami, R. TD430.W3634 2010 628--dc22 2010025803

Published by Nova Science Publishers, Inc. † New York

Water Production and Wastewater Treatment, Nova Science Publishers, Incorporated, 2010. ProQuest Ebook Central,

CONTENTS vii 

Preface Chapter 1

Chapter 2

Copyright © 2010. Nova Science Publishers, Incorporated. All rights reserved.

Chapter 3

Chapter 4

Chapter 5

Treatment and Reuse of Wastewater From A Petrochemical Complex Noah I. Galil and Yael Levinsky  Activated Sludge Characterization: Extraction and Identification of Hydrolytic Enzymes Debora Nabarlatz, Frank Stüber, Josep Font, Agustí Fortuny, Azael Fabregat and Christophe Bengoa 

11 

DGGE and 16S rDNA Sequencing Analysis of Bacterial Communities in a Membrane Bioreactor for the Removal of Phenol from Oil Refinery Wastewater Fernanda R. Pinhati, Aline F. Viero, Eduardo M. Del Aguila, Ana Paula R. Torres, Joab T. Silva and Vânia M. F. Paschoalin 

27 

Biosorption of Cd (II) and Ni (II) from Aqueous Solutions by Cystoseira indica M. M. Montazer-Rahmati, P. Rabbani and A. Abdolali  

45 

Multicomponent Removal of Heavy Metals from Aqueous Solution Using Low-Cost Sorbents Hilda Elizabeth Reynel-Avila, Didilia Ileana Mendoza-Castillo,  Virginia Hernández-Montoya and Adrián Bonilla-Petriciolet  

Chapter 6

The Adsorption of Dyes on Waste Tyre Derived Activated Carbon O. S. Chan, C. W. Wong and G. McKay 

Chapter 7

Adsorption of Basic Dyes by Activated Carbon from Waste Bamboo L. S. Chan, W. H. Cheung, S. J. Allen and G. McKay 

Chapter 8

1 

Investigations on Arsenic Adsorption onto Dolomitic Sorbents Y. Salameh, M. N. M. Ahmad, S. J. Allen and G. M. Walker 

Water Production and Wastewater Treatment, Nova Science Publishers, Incorporated, 2010. ProQuest Ebook Central,

69 

101 

117  133 

vi Chapter 9

Chapter 10

Chapter 11

Chapter 12

Copyright © 2010. Nova Science Publishers, Incorporated. All rights reserved.

Chapter 13

Contents Ozonation and Adsorption of MTBE in the Presence of Perfluorooctyl Alumina as a Catalyst or Adsorbent Amanollah Ebadi, Jafar Soltan and Sirous Shafiei  Photocatalytic Degradation of Phenolic Contaminants Using Titanium Dioxide Nano-Particles: Statistical Modeling and Process Optimization using p-Cresol S. Ray and J. A. Lalman 

179 

Kinetic and Reactor Modeling for the Degradation of Phenol in Water by UV/H2O2 Masroor Mohajerani, Mehrab Mehrvar and Farhad Ein-Mozaffari 

197 

Statistical Optimization of Reactive Blue 221 Decolorization by Fungal Peroxidise Hamid-Reza Kariminia and Vajihe Yousefi 

215 

Arsenic Contamination in Groundwater in Latin America: The Challenge of Providing Safe Drinking Water in the Developing World María Fidalgo de Cortalezzi, Fernando Yrazu and Paola Sabbatini 

Chapter 14

Efficacy Of TiO2 Doped with Copper for Water Disinfection B. Antizar-Ladislao, L. Wu and M. A. Khraisheh  

Chapter 15

Assessing Reverse Osmosis and Ion Exchange for Condensates Recycling in Fermentation Claire Fargues, Marjorie Gavach, Marielle Bouix and Marie-Laure Lameloise 

Chapter 16

159 

Application of Theoretical Scaling Potential Index to Predict Onset of Composite Calcium Carbonate and Calcium Sulfate Fouling and Crystal Types and Phases in Seawater Reverse Osmosis Treatment R. Sheikholeslami, Y. Wang and H. Yu 

Index

Water Production and Wastewater Treatment, Nova Science Publishers, Incorporated, 2010. ProQuest Ebook Central,

225  243 

253 

269  285 

Copyright © 2010. Nova Science Publishers, Incorporated. All rights reserved.

PREFACE This book includes selected paper and is the result of a very successful 3-day symposium on Water Production and Wastewater Treatment – Technologies, Advances and Issues as part of the 8th World Congress of Chemical Engineering in Montreal in August 2009 which my colleague, Dr. Blanca Antizar-Ladislao, and I organized and convened. The importance of the topic showed the interest, almost 100 paper and poster presentations, and number of people participated and contributed to the symposium. There were excellent talks covering various aspects and topical issues related to water production and wastewater treatment in industrial and municipal sectors. Some talks were focused on removal of a specific contaminant, specific technology for the process, experimental findings, or process modeling. Others had wider perspective and were covering the importance of water production and wastewater treatment in the context of our current industrial society and the energy and climate change. The papers covered a unique mix of issues related to both water production and wastewater treatment. Success of such events always depends on many people and factors. I would like to gratefully acknowledge the financial contributions of Royal Academy of Engineering for travel and attendance at the Congress. I would also like to thank all the participants and presenters at the Symposium for their contribution enriching the Symposium, my colleagues at the technical committee and reviewers for assisting in selection and review of papers, and Dr. Antizar-Ladislao for her significant contributions in co-organizing the Symposium with me.

Professor Roya Sheikholeslami, Ph.D., P.Eng., C.Eng., F.I.ChemE Chair of the Organizing Committee Symposium on Water Production and Waste Water Treatment 8th World Congress of Chemical Engineering

Water Production and Wastewater Treatment, Nova Science Publishers, Incorporated, 2010. ProQuest Ebook Central,

Copyright © 2010. Nova Science Publishers, Incorporated. All rights reserved. Water Production and Wastewater Treatment, Nova Science Publishers, Incorporated, 2010. ProQuest Ebook Central,

In: Water Production and Wastewater Treatment Editors: B. Antizar-Ladislao and R. Sheikholeslami

ISBN 978-1-61728-503-5 © 2011 Nova Science Publishers, Inc.

Chapter 1

TREATMENT AND REUSE OF WASTEWATER FROM A PETROCHEMICAL COMPLEX Noah I. Galil* and Yael Levinsky Faculty of Civil and Environmental Engineering Laboratory for Industrial Wastewater Treatment and Reuse Technion – Israel Institute of Technology Haifa 32000, Israel

ABSTRACT

Copyright © 2010. Nova Science Publishers, Incorporated. All rights reserved.

Wastewater from petrochemical complexes is characterized by a diversity of pollutants, including free and emulsified hydrocarbons, phenols, cresols, xylenols, sulfides, ammonia and cyanides. The treatment of this wastewater is usually based on a multiple stage approach, consisting of physical, chemical and biological treatment processes. Wastewater treatment and reuse has been developed and applied at a petrochemical complex in Haifa, Israel. The solution was based on: (a) multiple stage treatment, creating several technological barriers, in order to avoid uncontrolled emissions into the neighboring marine environment; (b) maximal reuse of treated effluent and oil, for minimizing the disposal of pollutants outside the industrial zone; (c) step-bystep development, design and implementation of the treatment process enabled to establish the best operation and efficiency at the existing units and these could be used as starting conditions in the development of the next treatment stages; (d) flexibility and complete independent operation of the treatment units significantly increased the reliability of achieving a final effluent of high quality. The biological treatment process has been efficiently protected by preliminary flow regulation, to control hydraulic and pollutant loading. Additional protection of the biotreatment was achieved by the removal of free and emulsified oil by gravitational oily-water separators (API) followed by dissolved nitrogen flotation (DGF). Biotreatment is achieved by aerated ponds followed by a submerged biological contactor (SBC) for the removal of dissolved organics and for nitrification. Effluent polishing treatment is operated by chemically-enhanced sedimentation and by sand filtration. The treatment-recycle system in the petrochemical industry provides cost-effective solutions and high quality effluent to the recipient water *

[email protected].

Water Production and Wastewater Treatment, Nova Science Publishers, Incorporated, 2010. ProQuest Ebook Central,

2

Noah I. Galil and Yael Levinsky bodies. The approach of treatment-recycling serves as a trigger to the industrial management, in addition to regulatory requirements, to invest in water treatment facilities.

Copyright © 2010. Nova Science Publishers, Incorporated. All rights reserved.

INTRODUCTION Process and manufacturing plants usually consume large amounts of water in various operations of production, cleaning and rinsing. Wastewater contains pollutants that are frequently environmentally regulated. An effective way to minimize wastewater and pollutant emissions is to design wastewater recycling, so that the used water could be reused to a maximum extent in the same plant. Petrochemical complexes are producing large amounts of wastewater, which is characterized by a diversity of pollutants including free and emulsified hydrocarbons, phenol, cresols, xylenols, sulfides, ammonia, and cyanides. The production processes usually include distillation, catalytic cracking, visebreaking, oil and waxes, ethylene, sulfur recovery and other processes. Due to national or regional water shortage, which results in low fresh water consumption, as well as the variety of production processes, petrochemical wastewater in arid or semi-arid regions are characterized by high concentrations of pollutants. They include several periodical streams from gasoline, kerosene and other products washeries, containing up to 12 percent phenols, most of them cresols and xylenols. Diwan et al. (1995) mentioned a great potential for recycling of effluents to solve water shortage for the industries, since in many cases the cost of treatment is modest compared to overall benefits. Asano et al. (1996) mentioned the status of national policies on wastewater treatment, wastewater reuse characteristics and some wastewater reuse experiences in Japan. Au et al. (1996) reported a great economic efficiency obtained by the use of a low cost filtration system working on petrochemical secondary effluent. Wijesinghe et al. (1996) reported a study based on the use of secondary effluent as cooling water makeup for inland industry in Australia. Brown and Mountain (1998) reported findings regarding general feasibility of wastewater reuse as cooling tower makeup at power plants in Maryland, USA. Buhrmann et al. (1999) used a spiral reverse osmosis plant to treat mine water and spent cooling water producing a new source of water for a power station. Angelakes et al. (1999) presented the status of wastewater reclamation and reuse around the Mediterranean basin and discussed existing guidelines and regulations, also presenting the possibility of developing uniform wastewater reuse standards. The potential for the recovery and reuse of cooling water in Taiwan has been reported by Shu-Hai et al. (1999). A brief overview of the reuse of treated industrial wastewater in cooling water systems is provided by Phulwar et al. (1999), including a case study of the reuse of treated effluent as cooling water at a refinery process plant in India. Large wastewater reuse projects in the UK, based on longterm international operation experience on reuse projects for the petrochemical, power and paper industries are discussed by Durham (2000). Yang et al. (2000) introduced a mathematical approach to design an optimal network when multiple pollutants are contained and the treated effluent can be reused to a maximum extent in the same plant. Zhong and Lai (2009) reported the reuse of effluent from a petrochemical company for make up to the cooling system. Wong (2000) described the pilot testing and implementation of a major advanced wastewater reclamation project to recover secondary effluent from a

Water Production and Wastewater Treatment, Nova Science Publishers, Incorporated, 2010. ProQuest Ebook Central,

Treatment and Reuse of Wastewater from A Petrochemical Complex

3

Copyright © 2010. Nova Science Publishers, Incorporated. All rights reserved.

municipal plant and blowdown from a cooling tower for reuse in a large petroleum plant. Durham (2000) reported about large wastewater reuse projects in the UK based on long term international operation experience on reuse projects for the petrochemical, power and paper industries. In the field of wastewater treatment processes membrane technology, especially membrane bioreactor (MBR) is being applied (Llop et al., 2009, Fratila-Apachitei et al., 2001, Galil et al., 2009). This chapter deals with a research and development project, which was carried out at a petrochemical complex located at a distance of about two miles from the Mediterranean coast in the Gulf of Haifa, Israel. The program included characterization of the wastewater main stream, as well as lateral streams generated by specific production processes (Galil et al., 1988). Laboratory and pilot plant studies on flocculation-dissolved air flotation (Galil and Wolf, 2000) enabled the design and operation of a full-scale treatment plant. A comparative study of three alternative biological processes: activated sludge, rotating biological contactor and aerated ponds provided the data for a biological treatment process based on two aerated lagoons in series, accomplished by a lime softening-clarification chemical plant (Galil and Rebhun, 1990; Galil and Rebun, 1991). A survey of the biological process occurring in the recirculated cooling system of the industrial complex enabled to operate this system as the recipient of the treated effluent, as well as a polishing nitrification bioreactor (Rebhun and Engel, 1988). Following the research results and conclusions, the full scale developed solution for treatment and reuse of petrochemical wastewater was based on: (a) multiple stage treatment, achieved by combining physical, chemical and biological processes, creating several technological barriers in order to avoid uncontrolled emissions into the neighboring river and marine environment; (b) maximal recycling of treated effluent and oil, for minimizing disposal of pollutants outside the industrial zone (Galil and Rebhun, 1992).

BASIC CONCEPTS The implementation of environmental quality regulations, regarding the disposal of effluent to the environment, usually to water bodies, is imposing careful considerations. By lowering the level of pollutants to the values required by the regulations, the treated effluent and some of the constituents separated from the wastewater could be considered for recycling by the petrochemical complex. This would minimize the disposal outside the industrial zone. In the case of the Haifa petrochemical complex, the research and development project included the following tasks: (a) characterization of the main raw wastewater streams, as well as lateral streams generated by specific production processes; (b) feasibility studies of general treatment of all the wastewater streams versus separate treatment of concentrated streams; the investigated process was based on chemical emulsion-breaking, flocculation and dissolved air flotation (DAF); (c) a comparative study of three alternative biological treatment processes for the removal of dissolved organic matter; (d) a survey of the processes occurring in the water cooling system of the complex, including studies on the use of treated effluent as makeup; (e) characterization of two different types of sludge produced by the wastewater treatment and development of sludge treatment methods.

Water Production and Wastewater Treatment, Nova Science Publishers, Incorporated, 2010. ProQuest Ebook Central,

4

Noah I. Galil and Yael Levinsky

Copyright © 2010. Nova Science Publishers, Incorporated. All rights reserved.

Figure 1. General description of the wastewater treatment-reuse system.

Following the conclusions of the research and development project, the wastewater facilities of the Haifa petrochemical complex include: separate storage and treatment of the concentrated phenol (spent soda) streams; storage and flow regulation (equalization) of the main raw wastewater stream; gravitational oil-water separator (OWS); chemical flocculation and dissolved gas (nitrogen) flotation (DGF); biological treatment for carbonaceous substrate removal; biological treatment by submerged bio-contactors (SBC) for nitrification; chemical precipitation for softening and clarification; sand filtration; effluent reuse as makeup in the water cooling system; blow-down treatment by chemical precipitation-sedimentation before the final disposal to the river; sludge collection, treatment and disposal. A general description of the wastewater treatment-reuse system is described in Figure 1.

DESCRIPTION OF TREATMENT UNITS Flow Regulation: The main wastewater stream could be influenced by factors such as rain floods and spills caused by unexpected accidents at the production units. For minimizing these influences, two flow regulation tanks with a total capacity of 45,000 m3 were built and connected to the system. The operation of these tanks, having a capacity of about five days of maximal flow, enables the operators to avoid sudden hydraulic or pollutant surges on the treatment units. Concentrated wastewater streams: Studies carried out by Galil and Rebhun (1988) indicated severe disturbances and inhibition caused by phenols included in the spent soda streams coming from the gasoline and other product washeries. As part of the general project, the concentrated spent-soda wastewater streams were separated from the sewerage system, stored in special tanks and gradually treated. The treatment is based on neutralization and separation between: gases, which are conducted to the flare; an oily phase including the phenols is recycled to the production processes; a water phase containing mainly inorganic salts is drained to wastewater.

Water Production and Wastewater Treatment, Nova Science Publishers, Incorporated, 2010. ProQuest Ebook Central,

Copyright © 2010. Nova Science Publishers, Incorporated. All rights reserved.

Treatment and Reuse of Wastewater from A Petrochemical Complex

5

Chemical flocculation and flotation (DGF): Laboratory and pilot plant studies have been carried out, developing design and operational parameters for this process. The flotation unit is covered in order to avoid VOC's emissions, therefore nitrogen is used instead of air. The flotation systems consist on three parallel units with a capacity of 200 m3/hr each. The flocculant in use is a cationic polyelectrolyte in a dose of 7 to 10 mg/L. Later studies performed by Galil and Wolf (2000) on this wastewater indicated that the chemical flocculation - DGF could remove efficiently the emulsified phase, which could be aggregated and separated up to the surface. However, it was found that the process could also remove substantial amounts of dissolved organic matter, due to the hydrophobic characteristics of some of the substances, which could bind to the solid surfaces. Biological treatment: A comparative study has been carried out including activated sludge, rotating biological contactor (RBC) and aerated lagoons (Galil and Rebhun, 1990). These bioprocesses represent different concepts: activated sludge and RBC are considered as intensive processes, developing high concentrations of active biomass and high cell residence time (CRT), while aerated ponds are considered as a partial bioprocess, involving low biomass and low CRT values without biosolids recycling. The aerated ponds alternative was adopted because of the possibility of lowering the investment cost. It was clear that in this case, additional biotreatment would be necessary. This alternative was based on sharing the bioprocess tasks between: (a) the aerated ponds, performing carbonaceous substrate removal (two days detention time); (b) second stage biological treatment by submerged biological contactors (SBC), mainly for nitrification; The experience accumulated over the last ten years shows that this combination has achieved good and reliable biological treatment (Table 1). Chemical clarification: A chemical contact flocculation-clarification unit, designed for a flow of 600 m3/hr is operated for efficient separation of biosolids and clarification. A second identical unit works on the treatment of water from inside the cooling system (side stream treatment). Part of the side stream treated effluent goes back to the cooling system, while the remaining effluent is disposed off to the neighboring river (Figure 1). Both contact flocculation-clarification units are operated at pH values of 10.7 by addition of lime for enabling removals of calcium carbonate and magnesium hydroxide. The removal of suspended solids is being enhanced by the use of a cationic polyelectrolyte as aid coagulant. Filtration: A gravity sand filtration unit operates after chemical enhancement by a cationic polyelectrolyte. The effluent fits all the quality requirements for being reused as make up in the cooling system or for being discharged to the river. In the future the filtration effluent will be treated by ultra filtration, reverse osmosis and reused as make up to steam production for the local power station (Figure 1). Sludge treatment: The petrochemical complex wastewater treatment system is producing two categories of sludge: (a) oily sludge is produced by oily water separators, by the dissolved gas flotation and also includes sediments from crude oil storage tanks. The oily sludge is gravitationally thickened in long term concrete storage tanks, chemically conditioned and cake-filtered by geo-tubes. The oil is recycled to production and the water phase is returned to wastewater treatment. (b) sludge produced by the biological treatment stages (biosolids) is obtained from the chemical clarification and from the backwash of the sand filters. This sludge is stabilized by land farming and transported to landfill sites. In the future all sources of sludge will be treated by cake filtration and by thermal technologies (Figure 2) for maximizing recyclable materials and improve land utilization.

Water Production and Wastewater Treatment, Nova Science Publishers, Incorporated, 2010. ProQuest Ebook Central,

6

Noah I. Galil and Yael Levinsky

Figure 2. General description of the sludge treatment system.

EFFLUENT QUALITY Table 1 presents the results obtained as part of a six years (2003 to 2008) monitoring of the last wastewater treatment stages, which are DGF, aerated ponds, submerged biological contactor and chemical clarification followed by sand filtration. The results are expressed in statistical terms which include 50% and 80% probabilities of obtaining values equal to or less than the stated magnitudes. Table 1 also indicates average values and standard deviation. The parameters reported include pH, total organic carbon, total suspended solids, oil, ammonia nitrogen, nitrate nitrogen and phosphates.

Copyright © 2010. Nova Science Publishers, Incorporated. All rights reserved.

Table 1. Effluent quality after different treatment stages Statistical pH TOC TSS Parameter (mg/L) (mg/L) DGF EFFLUENT 50% 7.6 59 80% 7.8 81 Average 7.6 53 St. Dev. 0.3 27 AERATED PONDS EFFLUENT 50% 7.8 28 45 80% 8.0 43 68 Average 7.9 33 51 St. Dev. 2.6 17 27 SUBMERGED BIOCONTACTOR EFFLUENT 50% 8.0 14 37 80% 8.1 19 56 Average 7.9 17 42 St. Dev. 0.2 11 24 SAND FILTRATION EFFLUENT 50% 8.1 5.7 5.6 80% 8.2 7.9 8.2 Average 7.9 6.8 6.4 St. Dev. 0.2 3.2 3.8

OIL (mg/L)

NH4-N (mg/L)

NO3-N (mg/L)

PO4 (mg/L)

16 25 23 14

9.5 14.0 10.9 8.7

0.9 2.5 1.7 2.8

2.0 2.8 2.0 0.9

7.4 13.7 11.4 18.4

6.3 9.5 6.8 4.9

0.8 2.7 2.5 4.7

0.7 1.4 1.0 1.8

4.1 7.6 6.4 11.8

0.1 0.2 0.5 2.0

32.2 42.0 32.9 14.7

1.3 2.0 1.5 0.9

1.8 3.4 2.3 0.8

0.1 0.2 0.4 0.7

28.7 37.1 31.7 15.8

0.6 0.8 0.7 0.4

Water Production and Wastewater Treatment, Nova Science Publishers, Incorporated, 2010. ProQuest Ebook Central,

Treatment and Reuse of Wastewater from A Petrochemical Complex

7

The final effluent, after sand filtration, contains less than 10 mg/L of TOC, less than 10 mg/L of suspended solids, less than 1 mg/L of ammonia nitrogen and less than 1 mg/L phosphates. This effluent is being reused as make up to the water cooling systems as well as for fire fighting. In the future part of the effluent will be additionally treated and reused for steam production (Figure 1).

CONCLUSIONS The project involves several technological barriers for protecting river and sea water and for enabling sustainable effluent reuse: • •

Copyright © 2010. Nova Science Publishers, Incorporated. All rights reserved.

•

hydraulic barrier is achieved by the storage-flow regulation tanks; physical and chemical barriers include four different treatment stages of gravity separation, chemically enhanced flotation, chemical clarification and sand filtration before reuse or final disposal; biological barriers include two separate processes: aerated ponds and submerged biological contactors.

Flexibility, as well as complete independent operation of the treatment units, significantly increased the reliability of producing a final effluent of high and reliable quality. The recycling of oil and sludge has minimized the disposal of contaminants outside the industrial zone. The treatment-recycle system in the reported petrochemical industry provides costeffective solutions. The reuse of treated effluent at the Haifa petrochemical complex saves about 2.5 million cubic meters of fresh water per year, which is the equivalent water consumption of a town with a population of 45,000. The cost of this amount of water, if purchased from the national resources would be close to one million US dollars per year. The cost-effective approach of treatment-recycling serves as a trigger to industrial management, in addition to the regulatory requirements, to invest in water treatment facilities.

ACKNOWLEDGMENT The project was sponsored by the Oil Refineries, Haifa Ltd., and carried out at the Laboratory for Industrial Wastewater Treatment and Water Renovation at the Faculty of Civil Engineering, Technion - Israel Institute of Technology. During the years 1979 - 1997, Professor Menahem Rebhun was the head of the above Laboratory and had a leading role in this project.

REFERENCES Asano, T., Maeda, M. and Takaki, M. (1996). Wastewater reclamation and reuse in Japan: Overview and implementation examples. Water Science Technology 34(11), 219-226.

Water Production and Wastewater Treatment, Nova Science Publishers, Incorporated, 2010. ProQuest Ebook Central,

Copyright © 2010. Nova Science Publishers, Incorporated. All rights reserved.

8

Noah I. Galil and Yael Levinsky

Angelakes, A.H., Marekos, DoMonte, M.H.F., Bontaoux L. and Asano, T. (1999). Water Research 33(10), 2201-2217. Au, D., Zhang, J. and Yi, Y. (1996). Using bundle filters to process petrochemical secondary effluents for industrial reuse. Water Science Technology 34(10), 127-131. Brown, D.H. and Mountain, D. (1998). Wastewater reuse as cooling tower makeup: A pioneering case study in Maryland. Proc. Annual Meet., Air Waste Mang. Assoc., 91st, CODEN: PAMEE5 ISN, 1052-6102. Buhrmann, F., Van der Waldt, M., Hanekom, D. and Finlayson, F. (1999). Treatment of industrial wastewater for reuse, S. Afr. Desalination 124(1-3), 263-269. Diwan, R.C., Bausal, T.K., Garg, M.R., Vilaj, K.and Tiwana, H.S. (1995). Reuse of wastewater for industries. Environ. Pollut. Prot., Deep and Deep Publications, New Delhi, India, 109-117. Durham, B. (2000). Case studies of wastewater reuse for the petrochemical, power and paper industry. Spec. Publ. R. Soc. Chem. (Membrane Technology in Water and Wastewater Treatment), 241-247. Fratila-Apachitei, L.E., Kennedy, M.D., Linton, J.D., Blume, I. and Schippers, J.C. (2001), Influence of membrane morphology on the flux decline during dead-end ultrafiltration of refinery and petrochemical waste water. Journal of Membrane Science 182(1-2), 151159. Galil, N.I., Rebhun, M. and Brayer, Y. (1988). Disturbances and inhibition in biological treatment of wastewater from an integrated oil refinery. Water Science Technology 20(10),21-29. Galil, N.I. and Rebhun M. (1990). A comparative study of RBC and activated sludge in biotreatment of wastewater from an integrated oil refinery. Proc. of the 44th Ind. Waste. Conf., Purdue Univ., West Lafayette, 711-717. Galil, N.I. and Rebhun, M. (1991). Combined treatment by aerated ponds and chemical clarification completed by recirculated cooling systems, Proc. 45th Ind. Waste Conf., Purdue Univ., West Lafayette, 645-654. Galil, N.I. and Rebhun M. (1992). Multiple technological barriers combined with recycling of water and oil in wastewater treatment of a coastal petrochemical complex. Water Science Technology 25 (12), 277-282. Galil, N.I. and Wolf, D. (2000). Removal of hydrocarbons from petrochemical wastewater by dissolved air flotation. 4th Int. Conf. on Dissolved Air Flotation, Helsinki, Finland. Galil, N.I., Ben-David Malachi, K. and Sheindorf, Ch. (2009). Biological nutrient removal by MBR configurations. Environmental Engineering Science, 26 (4), 817-824. Llop, A., Pocurull, E. and Borrull, F. (2009). Evaluation of the Removal of Pollutants from Petrochemical Wastewater Using A Membrane Bioreactor Treatment Plant. Water, Air, and Soil Pollution 197(1-4), 349-359. Phulwar, D. and Amesur, D. (1999). Reuse of wastewater in cooling water system. Trans. Met. Finish. Assoc., India 8 (1), 31-34. Rebhun, M. and Engel, G. (1988). Reuse of wastewater for industrial cooling systems, Jour. Water Poll. Control Fed. 60, 232-242. Shu-Hai, Y., Dyi-Hwa, T., Gia-Luen, G. and Jyh-Jian, Y. (1999). Resources conservation and recycling 26 (1), 53-70.

Water Production and Wastewater Treatment, Nova Science Publishers, Incorporated, 2010. ProQuest Ebook Central,

Treatment and Reuse of Wastewater from A Petrochemical Complex

9

Copyright © 2010. Nova Science Publishers, Incorporated. All rights reserved.

Wijesinghe, B., Kaye, R., Fell, C. and Joseph, D. (1996). Reuse of treated sewage effluent for cooling water makeup: a feasibility study and a pilot plant study. Water Science Technology 33 (10-11), 363-369. Wong, J. M. (2000). Testing and implementation of an advanced wastewater reclamation and recycling system in a major petrochemical plant. Water Science and Technology 42(5-6), 23-27. Yang, M.D., Sykes, R.M. and Merry, C.J. (2000). Estimation of algal biological parameters. Ecological Modelling 125(1), 1-13. Zhong, X. and Lai, X. (2009). Commercial application of properly treated wastewater discharged from ethylene plant. China Petroleum Processing and Petrochemical Technology (3), 25-28.

Water Production and Wastewater Treatment, Nova Science Publishers, Incorporated, 2010. ProQuest Ebook Central,

Copyright © 2010. Nova Science Publishers, Incorporated. All rights reserved. Water Production and Wastewater Treatment, Nova Science Publishers, Incorporated, 2010. ProQuest Ebook Central,

In: Water Production and Wastewater Treatment Editors: B. Antizar-Ladislao and R. Sheikholeslami

ISBN 978-1-61728-503-5 © 2011 Nova Science Publishers, Inc.

Chapter 2

ACTIVATED SLUDGE CHARACTERIZATION: EXTRACTION AND IDENTIFICATION OF HYDROLYTIC ENZYMES Debora Nabarlatz*,a, Frank Stübera, Josep Fonta, Agustí Fortunyb, Azael Fabregata, and Christophe Bengoaa

Copyright © 2010. Nova Science Publishers, Incorporated. All rights reserved.

a Departament d’Enginyeria Química, Escola Tècnica Superior d’Enginyeria Química, Universitat Rovira i Virgili. Av. Països Catalans 26 (43007) Tarragona, Catalonia, Spain b Departament d’Enginyeria Química, EPSEVG, Universitat Politècnica de Catalunya. Av. Víctor Balaguer s/n (08800) Vilanova i la Geltrú, Barcelona, Catalonia, Spain

ABSTRACT The organic matter in domestic wastewater is mainly composed by lipids, proteins and carbohydrates. Due to that the microorganisms present in activated sludge cannot consume it directly, they produce specific hydrolytic enzymes like proteases and lipases to metabolize the organic fraction. In this sense, the recovery of enzymes from activated sludge is a promising option for the valorisation of this resource, considering the sludge as a raw material capable of generating multiple new biomolecules and products. This chapter deals with the extraction and identification of the several types of hydrolytic enzymes that have been detected in activated sludge. The enzymes quantified were lipase, protease, alkaline and acid phosphatase, L-leu-aminopeptidase, α-glucosidase and αamylase. The amount of protein released during the treatments was also measured. The method employed for the extraction of enzymes from activated sludge was ultrasonication. The results showed that 30 min of disintegration applying 50 W of ultrasonic power are enough to extract the maximal amount of enzymatic activity. These results were compared with those obtained when 2% v/v Triton X100 (a detergent *

Corresponding author. Tel. +34 977558656, Fax. +34 977559667. E-mail: [email protected].

Water Production and Wastewater Treatment, Nova Science Publishers, Incorporated, 2010. ProQuest Ebook Central,

12

Debora Nabarlatz, Frank Stüber, Josep Font et al. currently used to extract membrane proteins) was used as additive. It was observed that the use of Triton X100 increases the amount of protease and acid phosphatase activity extrated, but it has not a significant influence in the extraction of the other enzymes.

Copyright © 2010. Nova Science Publishers, Incorporated. All rights reserved.

INTRODUCTION Nowadays, the activated sludge generated during wastewater treatment is becoming a serious issue due to several factors like the growth of population, its accumulation in large cities, and the increment in the amount and complexity of the industrial activity. As results from the implementation in 2005 of the Urban Waste Water Treatment Directive 91/271/EEC, the generation of sewage sludge has increased significantly. In Europe, the total amount of sludge generated in urban wastewater treatment plants has increased from 5.5 million (1992) to 10 million tons dry matter in 2007. In Spain, sewage sludge production from 1997 to 2005 increased 40%, rising up to 1.065 million tons in 2006 [1]. As consequence, the amount of activated sludge that is disposed off increases annually and several countries in Europe are implementing a more strict environmental legislation which considers not adequate the actual disposal ways (landfilling and composting). For this reason it is necessary to develop new technologies, firstly to reduce the amount of activated sludge generated that should be disposed off, and secondly to use it as raw material capable to generate biomolecules and energy. The recovery of valuable products from sludge that could be used for the sludge reduction itself or for any other industrial applications are promising. The domestic wastewater is mainly composed by complex organic matter, from which 60 - 70% is formed by lipids, proteins and polysaccharides. A huge fraction of this organic matter (30 - 85%) is formed by particles larger than 0.1 µm which cannot be directly assimilated by the microorganisms [2]. For this reason, the microorganisms produce hydrolytic enzymes that are released to the media to degrade this organic matter. Up to 8090% of the microorganisms present in activated sludge cannot be grown using standard cultivation techniques [3] and for this reason several methods have been applied to isolate the hydrolytic enzymes that are produced during the degradation. It was demonstrated that the enzymatic activity (as free form) present in the liquid phase of the activated sludge is almost negligible, and that the extracellular enzymes are attached to the flocs components (e.g. the cell wall of individual cells) by ionic and also hydrophobic interactions [3]. Ultrasound disintegration, alone or combined with detergents or ionic exchange resins is one of the methods with better performance for the recovery of the enzymes maintaining them active [2, 4-11]. The extraction of enzymes from activated sludge collected from municipal wastewater treatment plants was just briefly explored until now, as demonstrated by the fact that less than 15 articles were found to deal with this subject until now. The extraction of enzymes from activated sludge was started in 1995 [4]. In this study it was detected the presence of extracellular enzyme activity of esterase, lipase, leucine aminopeptidase, α- and βglucosidase, chitinase, and β-glucuronidase. Several authors reported these and other enzymes (13 types in total, overviewed in Table 1) as they are present in activated sludge collected from urban wastewater treatment plants. However, there is a wide range in the amount of enzyme activity recovered that can vary up to three orders of magnitude depending on the source.

Water Production and Wastewater Treatment, Nova Science Publishers, Incorporated, 2010. ProQuest Ebook Central,

Activated Sludge Characterization

13

Table 1. State of the art for the different enzymes extracted from activated sludge in municipal wastewater treatment plants Enzyme Esterase

Lipase

L-leuaminopeptidase

Protease Protease (continuation)

α-glucosidase

Amount detected 0.035 140

Substrate

Unit definition

Author

μmol MUF/h/mg VS μmol p-nitrophenol/min/L

[4] [30]

0.005

4-methylumbelliferyl stearate p-nitrophenyl esters (chain length 4-6) 4-methylumbelliferyl stearate

μmol MUF/h/mg VS

[4]

0.002 350 1000 90

p-nitrophenyl-palmitate p-nitrophenyl-palmitate 4-methylumbelliferyl oleate p-nitrophenyl-palmitate

μmol p-nitrophenol/min/g MLSS μmol p-nitrophenol/min/g VSS μmol MUF/h/g TS μmol p-nitrophenol/L/h

[7] [3] [31] [32]

0.015

μmol MCA/h/mg VS

[4]

9 38

L-leucine-4-methyl-7coumarinylamide hydrochloride L-leu-p-nitroanilide L-leu-p-nitroanilide

[2] [32]

0.1 80

Azocasein, 440 nm Casein

μmol p-nitroaniline/min/g VS μmol p-nitroaniline/L/h 1 mg azocasein/h/mg MLVSS

2210 8 4000 28 4 9 2.5 0.005

μmol L-tyrosine/min/mg protein

[29] [6]

Casein Azocasein, 340 nm Azocasein, 440 nm Casein Casein Casein Casein

μmol L-tyrosine/min/g MLSS ΔAbs/min/g VS 0.01 ΔAbs/g VSS μmol L-tyrosine/min/g VSS μmol L-tyrosine/min/g VSS μmol L-tyrosine/min/g VSS μmol L-tyrosine/min/g VSS

[7] [2] [3] [8] [9] [10] [11]

4-methylumbelliferyl-α-Dglucoside p-nitrophenyl-α-Dglucopyranoside -

μmol MUF/h/mg VS

[4]

μmol p-nitrophenol/min/g VS

[2]

μmol glucose/min/g VSS μmol glucose/min/g VSS μmol glucose/min/g VSS

[8] [9] [10]

μmol MUF/h/mg VS

[4]

units/g MLSS

[7]

μmol p-nitrophenol/L/h

[32]

μmol MUF/h/mg VS

[4]

0.002 2.7

4-methylumbelliferyl-β-Dglucoside p-nitrophenyl-β-Dglucopyranoside p-nitrophenyl-β-Dglucopyranoside 4-methylumbelliferyl-β-Dglucuronide Carboxymethyl cellulose Amylose azure

units/g MLSS

[7] [2]

15 150 45 20

-

0.2

p-nitrophenyl phosphate

38 60

p-nitrophenyl phosphate p-nitrophenyl phosphate

Copyright © 2010. Nova Science Publishers, Incorporated. All rights reserved.

2.5 300 4 2 β-glucosidase

0.009 0.002 40

β-glucuronidase 0.005 Endoglucanase α-amylase

Alkaline phosphatase

μmol Remazol brilliant blue/min/g VS μmol glucose/min/g VSS μmol glucose/min/g VSS μmol glucose/min/g VSS μmol glucose/min/g VSS μmol p-nitrophenol/h/ mg MLVSS μmol p-nitrophenol/L/h μmol p-nitrophenol/h/ mg MLVSS

Water Production and Wastewater Treatment, Nova Science Publishers, Incorporated, 2010. ProQuest Ebook Central,

[8] [9] [10] [11] [29] [32] [9]

Debora Nabarlatz, Frank Stüber, Josep Font et al.

14

Table 1. (Continued) Enzyme

Amount detected 17

Substrate

Unit definition

Author

p-nitrophenyl phosphate

[10]

10

p-nitrophenyl phosphate

μmol p-nitrophenol/h/ mg MLVSS μmol p-nitrophenol/h/ mg MLVSS

0.5

p-nitrophenyl phosphate

[29]

38

p-nitrophenyl phosphate

12

p-nitrophenyl phosphate

3

p-nitrophenyl phosphate

μmol p-nitrophenol/h/ mg MLVSS μmol p-nitrophenol/h/ mg MLVSS μmol p-nitrophenol/h/ mg MLVSS μmol p-nitrophenol/h/ mg MLVSS

Dehydrogenase

0.35

Iodonitrotetrazolium (INT)

μmol INT-formazan /h/mg MLVSS

[29]

Chitinase

0.005

4-methylumbelliferyl-N-acetyland-D-glucosaminide

μmol MUF/h/mg VS

[4]

Copyright © 2010. Nova Science Publishers, Incorporated. All rights reserved.

Acid phosphatase

[11]

[9] [10] [11]

The definition of enzymatic activity is arbitrary and depends strongly on the substrate and the conditions of the reaction used to determine it. For this reason, it is necessary to establish formal protocols with well defined methods (in which the definition of enzymatic activity is clear) suitable to determine the enzymatic activities, because this is a key parameter in the extraction and purification process. The economy of an industrial process for the extraction of enzymes will be determined by the efficiency of the extraction process and the enzymatic activity extracted. About the possible applications of the extracted enzymes, several studies have been carried out to evaluate the influence of an enzymatic pretreatment step prior to the anaerobic digestion of domestic or industrial wastewater. The results showed that it was possible to remove solids, decrease the COD level and improve the biogas production after the anaerobic digestion. But, for these studies, mainly commercial enzymes (mostly lipases) have been used [12-22] and just in few cases enzymes extracted from sludge were used to treat and improve the degradation of different compounds in certain types of wastewater [23]. On the other side, enzymes such as lipases are starting to be used as a pretreatment step prior to the anaerobic degradation of specific types of polymeric esters like phthalate esters [24], diethyleneglycol terephtalate and poly(ethylene)terephtalate [23]. These studies open the possibilities of application for the different enzymes recovered from activated sludge. This will have two main advantages: first, the recovery of a valuable product from sludge; and second, that these enzymes could be used to improve the degradation of sludge (which will enhance the biogas production during the anaerobic digestion step) or for the degradation of other specific chemical compounds that are difficult to degrade by traditional processes. The objective of the present chapter is to determine the different types of hydrolytic enzymes and the amount of enzymatic activity present in the activated sludge of an urban wastewater treatment plant, using ultrasound disintegration alone or combined with a nonionic detergent like Triton X100. The protocols for the determination of the enzymatic activity recovered and the definition of the enzymatic activity itself will be established.

Water Production and Wastewater Treatment, Nova Science Publishers, Incorporated, 2010. ProQuest Ebook Central,

Activated Sludge Characterization

15

MATERIAL AND METHODS Sludge Collection and Handling The activated sludge was collected from the aeration basin in the municipal wastewater treatment plant in Reus, Spain. This plant processes near 20,000 m3 wastewater/day (the city has 105,000 inhabitants). The samples of activated sludge were taken and transported to the laboratory in 30 min. All the disintegration experiments were carried out the same day, preserving the sludge under aeration at room temperature.The samples of sludge were taken every 2-3 weeks, and the experiments were carried out along 6 months of plant operation. The sludge was used as received, and it was analyzed in order to determine the content in total solids (TS) and volatile suspended solids (VSS) by gravimetric method according to standard methods [25].

Copyright © 2010. Nova Science Publishers, Incorporated. All rights reserved.

Disintegration of Activated Sludge The disintegration experiments were carried out using an ultrasonic disintegrator UP200S (Hielscher Ultrasonics GmbH, Germany). The experiments were carried out at 24 kHz working frequency, 4 W/cm2 power intensity (50 W ultrasonic power) and different disintegration times between 1 and 30 min. The volume of sludge used for each test was 200 mL using 400 mL vessels. The disintegration experiments were carried out using sludge alone or combined with a non ionic detergent like Triton X100 (TX100) in concentration 2% v/v (purchased from Sigma Aldrich). These conditions were chosen taking into account preliminary experiments for the recovery of protease and lipase [26, 27]. For all the experiments a water–ice bath was used, maintaining the temperature constant at 5±1ºC. Samples of sludge were taken before and after the disintegration process and were centrifuged at room temperature at 10,100 × g for 10 min prior to the analysis. The supernatant was used as the source of enzyme for the determination of the enzymatic activity and the amount of protein. All the disintegration experiments were carried out by duplicate.

Determination of the Enzymatic Activity and Amount of Protein The enzymes identified and quantified were lipase, protease, alkaline phosphatase, acid phosphatase, L-leu-aminopeptidase, α-glucosidase and α-amylase. All the methods used to determine the enzymatic activity were based on standard protocols from Sigma Aldrich. The amount of protein was quantified following the Bradford method. Absorbance measurements were carried out using an UV-Vis DINKO spectrophotometer (DINTER S.A. Spain). All the experiments were carried out by duplicate. o

Lipase activity. It was determined using p-nitrophenyl butyrate as substrate, measuring the release of p-nitrophenol by continuous spectrophotometric rate determination. For the experiments, 0.2 mL of enzyme solution were placed in a suitable cuvette, adding 1.8 mL of a solution containing 100 mM NaH2PO4 buffer

Water Production and Wastewater Treatment, Nova Science Publishers, Incorporated, 2010. ProQuest Ebook Central,

Debora Nabarlatz, Frank Stüber, Josep Font et al.

16

o

Copyright © 2010. Nova Science Publishers, Incorporated. All rights reserved.

o

o

o

having 150 mM NaCl and 0.5% v/v TX100 (pH 7.2 at 37ºC). It was mixed by inversion and the absorbance at 400 nm was measured at room temperature. The reaction started when 20 μL of 50 mM p-nitrophenyl butyrate (PNPB) prepared in acetonitrile were added to the test, and 20 μL of H2O were added to the blank. The increase in the absorbance at 400 nm was recorded for approximately 6 min. The ΔAbs400nm/min was calculated using the maximum linear rate for both the test and the blank, considering that 0.0148 is the µmolar extinction coefficient of p-nitrophenol at 400 nm. One unit of enzyme will release 1.0 μmol of p-nitrophenol per minute. Protease activity. It was analyzed using casein as substrate, measuring by spectrophotometry the concentration of L-tyrosine released by the action of the enzyme. A 0.65% w/v casein solution was prepared in 50 mM potassium phosphate buffer (pH 7.5 at 37ºC). 5 mL of this substrate solution were mixed with 1 mL of the enzyme solution and incubated during 10 min at 37ºC. Then 5 mL of 110 mM trichloroacetic acid were added to stop the reaction. After 30 min of incubation at 37ºC, the reaction mixture was centrifuged at 10,100 × g for 10 min at room temperature. The clear supernatant (2 mL) was mixed with 5 mL of 500 mM Na2CO3 and 1 mL of Folin and Ciocalteu’s Phenol reagent. This mixture was incubated for 30 min at 37ºC and finally centrifuged at 10,100 × g for 10 min at room temperature. The absorbance of this solution was measured at 660 nm (1 cm light path cuvettes) against a blank. The concentration of L-tyrosine was determined by comparison with a calibration curve. The unit definition for the enzymatic activity considers that one unit of protease will hydrolyze casein to produce colour equivalent to 1.0 µmol of L-tyrosine per minute. Acid phosphatase activity. It was determined by spectrophotometric stop rate determination using p-nitrophenyl phosphate as substrate. For this analysis, 0.5 mL of 90 mM citrate buffer (pH 4.8 at 37ºC) and 0.5 mL of 15.2 mM p-nitrophenyl phosphate (PNPP) were added to test tubes (test and blank). The mixture was equilibrated at 37ºC, and then 0.10 mL of enzyme solution were added to the test tube. Immediately it was mixed by inversion and incubated at 37ºC for exactly 10 min. Then, 4 mL of 100 mM NaOH were added to the test and blank, and finally 0.10 mL of the enzyme solution were added to the blank. The absorbance was measured at 410 nm. The definition of enzymatic activity considers that one unit of enzyme hydrolyzes 1 µmol of p-nitrophenyl phosphate per minute at pH 4.8 at 37ºC. Alkaline phosphatase activity. It was determined using p-nitrophenyl phosphate as substrate, measuring the release of p-nitrophenol by continuous spectrophotometric rate determination. For this test, 1.35 and 1.4 mL of 1 M diethanolamine buffer with 0.5 mM MgCl2 (pH 9.8 at 37ºC) were placed in the test and blank cuvette, respectively. 0.15 mL of a 150 mM p-nitrophenyl phosphate solution were added to both cuvettes. The mixture was equilibrated at room temperature, and then 0.05 mL of the enzyme solution were added to the test. The increase in the absorbance at 405 nm was recorded for approximately 10 min. The definition of enzyme activity considers that one unit of enzyme hydrolyzes 1 µmol of p-nitrophenyl phosphate per minute at pH 9.8 at 37ºC. Leucine aminopeptidase activity. It was determined by continuous spectrophotometric rate determination using L-leucine-p-nitroanilide as substrate. To

Water Production and Wastewater Treatment, Nova Science Publishers, Incorporated, 2010. ProQuest Ebook Central,

Activated Sludge Characterization

o

Copyright © 2010. Nova Science Publishers, Incorporated. All rights reserved.

o

o

17

this purpose, 2.8 mL of 50 mM sodium phosphate buffer (pH 7.2 at 37ºC) were placed in suitable cuvettes (test and blank), and 0.1 mL of 24 mM L-leucine-p-nitroanilide (prepared in methanol) were added to both test and blank. The mixture was equilibrated at room temperature, and 0.10 mL of the enzyme solution and deionized water were added to the test and the blank, respectively. The increase in the absorbance at 405 nm was recorded for approximately 7 min. The definition of enzyme activity considers that one unit of enzyme hydrolyzes 1 µmol of L-leucine-p-nitroanilide per minute at pH 7.2 at 37ºC. α-amylase activity. It was determined by colorimetric method using starch as substrate. To this purpose, 1 mL of 1% w/v starch solution prepared in 20 mM sodium phosphate buffer having 6.7 mM NaCl (pH 6.9 at 20ºC) were placed in suitable test tubes (for test and blank). The mixture was equilibrated at 20ºC. Then 1 mL of enzyme solution was added to the test, and the mixture was incubated for exactly 3 min at 20ºC. After that, 1 mL of colour reagent solution was added to the test and the blank, and finally 1 mL of enzyme solution was added to the blank. The colour reagent solution was prepared adding 8 mL of 1.5 g/mL sodium potassium tartrate in 2 M NaOH solution, to 20 mL of 96 mM 3,5-dinitrosalicylic acid solution, and this mixture was diluted to 40 mL with deionized water. The tubes were placed in a boiling water bath for exactly 15 min, and then they were cooled on ice to room temperature. Finally 9 mL of deionized water were added to the test and blank, and the absorbance was measured at 540 nm. A standard curve was prepared using maltose as standard solution, plotting ΔA540 nm of the standards vs. mg of maltose. The definition of enzyme activity considers that one unit of enzyme will liberate 1 mg of maltose from starch in 3 min at pH 6.9 at 20ºC. α-glucosidase activity. It was determined by spectrophotometric stop rate determination using p-nitrophenyl α-D-glucoside as substrate. 5 mL of 67 mM potassium phosphate buffer (pH 6.8 at 37ºC) and 0.20 mL of 3 mM glutathione (reduced solution) were placed in test tubes (for test and blank). 0.20 mL of enzyme solution and 0.20 mL of deionized water were added to the test and blank tube, respectively. The mixture was equilibrated at 37ºC, and then 0.5 mL of 10 mM p-nitrophenyl α-D-glucoside were added to both (test and blank) tubes. Then, the tubes were incubated for exactly 20 min at 37ºC. After this time, 2 mL of the corresponding test and blank were placed into suitable containers, adding 8 mL of 100 mM sodium carbonate solution to each tube. The absorbance was measured at 400 nm. The definition of enzyme activity considers that one unit of enzyme will liberate 1 µmol of D-glucose per minute at pH 6.8 at 37ºC. Protein. The amount of protein was determined following the Bradford method (BioRad). For this assay, 100 μL of the enzyme solution (test) or deionized water (blank) were placed in suitable test tubes, adding 5 mL of Coomasie blue brilliant reagent (Bradford colour reagent, diluted 1:5 with deionized water). The solution was incubated at room temperature for 1 h and the absorbance was measured at 595 nm. The samples containing detergent TX100 were diluted 1:20 to have a detergent concentration compatible with the assay. A calibration curve using bovine serum albumin (BSA) as standard was carried out in the range 0.01 to 1 mg protein/mL.

Water Production and Wastewater Treatment, Nova Science Publishers, Incorporated, 2010. ProQuest Ebook Central,

Debora Nabarlatz, Frank Stüber, Josep Font et al.

18

Copyright © 2010. Nova Science Publishers, Incorporated. All rights reserved.

RESULTS AND DISCUSSION As it was already mentioned, the objective of this chapter is to determine the presence of several hydrolytic enzymes in activated sludge collected from an urban wastewater treatment plant. To this purpose, several disintegration experiments were carried out, following the optimal experimental conditions found in preliminary studies [26, 27]. Two enzymes were detected in these previous experiments (protease and lipase), which were the most abundantly referenced in the literature. The present chapter extends this study to another five enzymes which are well known of participating in the degradation process. As it was mentioned before, it is difficult to compare the results and extract useful conclusions from the literature. Between them, the lack of details about the definition of enzymatic activity itself and the differences in the protocol and the substrate used to measure the enzymatic activity are of importance. For this reason, with the purpose of extracting the enzymes and to use them for any specific application, it is necessary to perform a detailed study about the availability of each type of enzyme and the optimal protocol for its extraction, maximizing the amount of enzyme activity recovered in each case. The enzymes studied are acid phosphatase, alkaline phosphatase, L-leu-aminopeptidase, α-amylase, protease, lipase, and α-glucosidase. Its presence in the activated sludge give us an idea of the variations in the substrate composition, microbial population, etc., depending on the presence of the different macromolecules in the activated sludge that the microorganisms are able to degrade [28]. Lipase hydrolyzes the lipids and esters present in activated sludge, while acid and alkaline phosphatase hydrolyze different types of phosphate esters. The generic enzyme called protease hydrolyzes different types of proteins, while leucine aminopeptidase is more specific and cuts the bond between leucine and different peptides. Finally, α-amylase and α-glucosidase hydrolyze starch and maltose from cellulose, respectively. The solids content in the sludge samples was determined by gravimetry according to standard methods. The results are showed in Table 2, together with the sludge characteristics. The total solids content varied between 1.8 – 2.2 g/L, while the volatile suspended solids (VSS) content varied between 0.87 – 0.92 g/L for all the samples tested. Table 2. Characteristics of the activated sludge collected from biological reactor Activated sludge from WWTP Reus TS (g/L)

2.03 ± 0.17

VSS (g/L)

0.90 ± 0.03

VSS/TS

0.44

Organic load (kg DBO5/ (kg TSS * day)

0.91 ± 0.10

Sludge age (days)

3

Water Production and Wastewater Treatment, Nova Science Publishers, Incorporated, 2010. ProQuest Ebook Central,

Copyright © 2010. Nova Science Publishers, Incorporated. All rights reserved.

Activated Sludge Characterization

19

The variations in the solids content are related to the typical variations in the organic matter composition of activated sludge due to changes in flow, temperature, seasonal changes, city population, etc. which at the end will affect the enzymatic activity content in the activated sludge. Figure 1 shows the results for the enzymatic activity obtained and the amount of protein recovered when using ultrasonication at different disintegration times. The disintegration experiments were carried out using a water-ice bath to control the temperature during the process (5±1ºC). The enzymatic activity of acid phosphatase, alkaline phosphatase, L-leuaminopeptidase, α-amylase, protease, lipase and α-glucosidase, is presented in units of enzyme per gram of VSS as well as in units of enzyme per mg of protein. The amount of protein released is also shown in Figure 1 h. As it can be observed, 30 min of disintegration time using 4 W/cm2 power intensity seems to be adequate to extract the maximal amount of enzyme present in the activated sludge for all the enzymes. However, there are some differences in the amount of the enzymatic activity extracted during the time of disintegration. The amount of enzymatic activity recovered per gram of VSS for acid and alkaline phosphatase, leucine aminopeptidase, α-glucosidase and lipase has an increasing tendency that seems to continue beyond 30 min of disintegration. On the other side, α-amylase and protease showed a maximal amount of enzymatic activity at 20 min of disintegration time, and beyond this point, the activity slightly diminishes. This decrease in the enzyme activity extracted could be explained by the degradation of the enzyme due to ultrasonic disintegration. Acid and alkaline phosphatase yielded a maximum of 7.8 and 3.4 units enzyme/g VSS (0.01 and 0.003 units/mL) respectively when using 30 min of disintegration time. These values are near to that found in the literature (0.0125 units/mL and 0.005 units/mL for acid and alkaline phosphatase, respectively) which were expressed in similar units [29]. Yu et al. reported between 10 – 17 units /g VSS for alkaline phosphatase and between 3 - 12 units/g VSS for acid phosphatase, respectively [10, 11]. L-leucine aminopeptidase showed a maximal activity of 11 units/g VSS, which is slightly higher than the value found in literature (around 9 units/g VSS) [2]. The results for α-amylase showed a higher enzymatic activity (it was possible to extract up to 174 units/g VSS) which is in the range reported (between 15 – 150 units/g VSS) [9-11]. The activity of α-glucosidase extracted after 30 min of disintegration (3.9 units/g VSS) was higher than some values found in the literature (2 units/g VSS) [2]. However, Yu et al. had extracted between 2 – 305 units/g VSS [8-10]. There are several results in the literature for the extraction of lipase, but only few of them are comparable. In the present study it was possible to extract up to 29.5 units/g VSS, which is a lower value than that found in literature (335 units/g VSS) [3]. On the other side, protease yielded up to 25.7 units/g VSS, which is in the range of the results found in the literature (between 2,5 – 28 units/g VSS) [8-11]. Finally, the maximal amount of protein present in the supernatant after centrifugation is around 0.035 mg/mL, which represents near 4% of the volatile suspended solids present in the activated sludge. If we observe the amount of enzyme recovered per mg of protein, it is possible to detect a maximum in the amount extracted at 10 min of disintegration, and then a steep decrease. However, observing the amount of protein extracted from activated sludge it increases until reaching a maximum at 20 min of extraction for all the cases (see Figures 1 a – g). That indicates that even if the amount of protein extracted increases with disintegration time, the effect of the ultrasonication affects the enzyme and deactivates it.

Water Production and Wastewater Treatment, Nova Science Publishers, Incorporated, 2010. ProQuest Ebook Central,

0 .3 0

10

Acid phosphatase (Units/g VSS)

(a )

0 .2 5

8

0 .2 0 6 0 .1 5 4 0 .1 0 2

0

0 .0 5

0

5

10

15

20

25

30

35

0 .0 0

Acid phosphatase (Units/mg protein)

Debora Nabarlatz, Frank Stüber, Josep Font et al.

20

0.14

(b)

0.12

4

0.10 3

0.08 0.06

2

0.04 1

0

0.02

0

5

10

15

20

25

0.00 35

30

Alkaline phosphatase (Units/mg protein)

5

16

0.5

(c)

14

0.4

12 10

0.3

8 0.2

6 4

0.1

2 0

0

5

10

15

20

25

Time (min)

Water Production and Wastewater Treatment, Nova Science Publishers, Incorporated, 2010. ProQuest Ebook Central,

30

35

0.0

Leucine aminopeptidase (Units/mg protein)

Time (min)

Leucine aminopeptidase (Units/g VSS)

Copyright © 2010. Nova Science Publishers, Incorporated. All rights reserved.

Alkaline phosphatase (Units/g VSS)

T im e (m in )

Activated Sludge Characterization

21 8

Amylase (Units/g VSS)

(d) 150

6

100

4

50

2

0

0

5

10

15

20

25

30

35

Amylase (Units/mg protein)

200

0

Time (min) 30

1.2 1.0

20

0.8

15

0.6

10

0.4

5

0.2

0

0

5

10

15

20

25

30

35

0.0

Time (min) 40

(f)

1.0

30 0.8 20

0.6 0.4

10 0.2 0

0

5

10

15

20

25

30

Time (min) Water Production and Wastewater Treatment, Nova Science Publishers, Incorporated, 2010. ProQuest Ebook Central,

0.0 35

Lipase (Units/mg protein)

1.2

Protease (Units/mg protein)

25

Lipase (Units/g VSS)

Copyright © 2010. Nova Science Publishers, Incorporated. All rights reserved.

Protease (Units/g VSS)

(e)

Debora Nabarlatz, Frank Stüber, Josep Font et al.

22

Glucosidase (Units/g VSS)

(g)

0.16

4

0.14 0.12

3

0.10 0.08

2

0.06 0.04

1

0.02 0

0

5

10

15

20

25

30

Glucosidase (Units/mg protein)

0.18

5

0.00 35

Time (min) 0.06

(h)

Copyright © 2010. Nova Science Publishers, Incorporated. All rights reserved.

Protein (mg/mL)

0.05 0.04 0.03 0.02 0.01 0.00

0

5

10

15

20

25

30

35

Time (min) Figure 1. Variation in the enzymatic activity of: (a) acid phosphatase; (b) alkaline phosphatase; (c) leucine aminopeptidase; (d) α-amylase; (e) protease; (f) lipase; (g) α-glucosidase; (h) amount of protein, as function of the disintegration time at the following conditions: 4 W/cm2 power intensity, 5±1ºC, 0% v/v TX100. Units of enzyme/g VSS; Ì Units of enzyme/mg protein.

According to our previous results [26, 27] the use of Triton X-100 (TX100) as additive in a concentration 2% v/v improves the process efficiency for the extraction of protease, but it has no effect in the extraction of lipase. For this reason, the disintegration experiments that were carried out without any additive (0% TX100) were compared with the results obtained when 2% TX100 was used. The disintegration experiments were done using 30 min disintegration time at 5±1ºC (see Figure 2 a). As it can be observed, only two enzymes incremented the activity recovered per gram of VSS: protease and acid phosphatase. The activity of protease increased from 11.7 to 53.3 enzyme units/g VSS (356% increase) when

Water Production and Wastewater Treatment, Nova Science Publishers, Incorporated, 2010. ProQuest Ebook Central,

Activated Sludge Characterization

23

0% and 2% of detergent was used, respectively. The activity of acid phosphatase increased from 15 to 20 enzyme units/g VSS at the same conditions (33% increase). The other enzymes were recovered with near the same activity when the experiments were carried out with or without detergent, indicating that the use of TX100 does not improve significantly its recovery. On the contrary, the activities of lipase or leucine aminopeptidase even descend with the use of TX100 (it decreases between 15-20%).

Enzyme activity (Units/g VSS)

180

(a)

Original 0% TX100 2% TX100

160 140 120 60 40 20 0

Enzyme activity (Units/mg protein)

Copyright © 2010. Nova Science Publishers, Incorporated. All rights reserved.

Lip. 15.0 12.5 10.0

Prot.

Acid

Alk.

Leu.

Amyl. Gluc.

(b)

Original 0% TX100 2% TX100

1.5

1.0

0.5

0.0 Lip.

Prot.

Acid

Alk.

Leu.

Amyl. Gluc.

Figure 2. Enzymatic activity recovered after disintegration experiments using 0% v/v and 2% v/v TX100 at 4 W/cm2 power intensity, 5±1ºC, 30 min disintegration time. (a) Enzymatic activity expressed in units/g VSS; (b) Enzymatic activity expressed in units/mg protein. Legends: Lip. (lipase); Prot. (protease); Acid (acid phosphatase); Alk. (alkaline phosphatase); Leu. (leucine aminopeptidase); Amyl. (α-amylase); Gluc. (α-glucosidase). Original (supernatant from sludge not disintegrated); 0% TX100 (supernatant from sludge disintegrated without TX100); 2% TX100 (supernatant from sludge disintegrated with 2% TX100).

Water Production and Wastewater Treatment, Nova Science Publishers, Incorporated, 2010. ProQuest Ebook Central,

24

Debora Nabarlatz, Frank Stüber, Josep Font et al.

The results in Figure 2 b showed the same tendencies. As the amount of protein extracted increases with the use of detergent, the amount of enzymatic activity decreases, indicating some deactivation of the enzyme. Only protease showed a little increment in the amount of enzyme extracted with the use of detergent. As it was already mentioned, the use of Triton X100 as additive will be effective depending on the type of enzyme that will be extracted. In this case its use is helpful for the extraction of protease and acid phosphatase, but it may generate problems related to the purification procedure that should be avoided in the subsequent steps of the process. For this reason, the extraction process should be optimized in every particular case for all the enzymes present in the activated sludge that will be recovered, taking into account the degree of purity and the specific enzymatic activity required at the end of the process.

CONCLUSION The main objective of this chapter was to analyze the different enzymes that take part in the degradation process of the activated sludge. Until now, in the literature there were few references about the enzymology of activated sludge, being difficult to compare them mainly due to the differences in the measurement protocols and the definition of enzymatic activity. This study allowed determining and quantifying the presence of the different hydrolytic enzymes in a standard way that could be measured, reproduced and compared in terms of enzyme units that could be useful for further applications.

Copyright © 2010. Nova Science Publishers, Incorporated. All rights reserved.

ACKNOWLEDGMENTS The financial support for this research was provided by the European Union Research 6th Framework Program, project REMOVALS, FP6-018525. Authors want to thank Jaume Cabré, Raúl García and Iñaqui Oriol from the company Gestió Ambiental i Abastament S.A. (EDAR de Reus, Spain) for their kind collaboration during this project.

REFERENCES [1]

[2]

[3] [4]

Carbonell, G.; Pro, J.; Gómez, N.; Babýn, M. M.; Fernandez, C.; Alonso, E.; Tarazona, J. V. Sewage sludge applied to agricultural soil: Ecotoxicological effects on representative soil organisms. Ecotoxicol. Environ. Saf. 2009, 72, 1309-1319. Cadoret, A.; Conrad, A.; Block, J. C. Availability of low and high molecular weight substrates to extracellular enzymes in whole and dispersed activated sludges. Enzyme Microb. Technol. 2002, 31, 179-186. Gessesse, A.; Dueholm, T.; Petersen, S. B.; Nielsen, P. H. Lipase and protease extraction from activated sludge. Water Res. 2003, 37, 3652-3657. Frolund, B.; Griebe, T.; Nielsen, P. H. Enzymatic Activity in the Activated-Sludge Floc Matrix. Appl. Microbiol. Biotechnol. 1995, 43, 755-761.

Water Production and Wastewater Treatment, Nova Science Publishers, Incorporated, 2010. ProQuest Ebook Central,

Activated Sludge Characterization [5]

[6]

[7]

[8] [9] [10]

[11] [12] [13]

[14]

Copyright © 2010. Nova Science Publishers, Incorporated. All rights reserved.

[15] [16]

[17] [18] [19]

[20] [21]

[22]

25

Frolund, B.; Palmgren, R.; Keiding, K.; Nielsen, P. H. Extraction of extracellular polymers from activated sludge using a cation exchange resin. Water Res. 1996, 30, 1749-1758. Jung, J.; Xing, X. H.; Matsumoto, K. Kinetic analysis of disruption of excess activated sludge by Dyno Mill and characteristics of protein release for recovery of useful materials. Biochem. Eng. J. 2001, 8, 1-7. Jung, J.; Xing, X. H.; Matsumoto, K. Recoverability of protease released from disrupted excess sludge and its potential application to enhanced hydrolysis of proteins in wastewater. Biochem. Eng. J. 2002, 10, 67-72. Yu, G.; He, P.; Shao, L.; Lee, D. Enzyme activities in activated sludge flocs. Appl. Microbiol. Biotechnol. 2007, 77, 605-612. Yu, G.; He, P.; Shao, L.; Lee, D. Extracellular enzymes in sludge flocs collected at 14 full-scale wastewater treatment plants. J. Chem. Technol. Biot. 2008, 83, 1717-1725. Yu, G.; He, P.; Shao, L.; Zhu, Y. Extracellular proteins, polysaccharides and enzymes impact on sludge aerobic digestion after ultrasonic pretreatment. Water Res. 2008, 42, 1925 - 1934. Yu, G.; He, P.; Shao, L.; Zhu, Y. Enzyme extraction by ultrasound from sludge flocs. J. Environ. Sci. 2009, 21, 204 - 210. Barjenbruch, M.; Kopplow, O. Enzymatic, mechanical and thermal pre-treatment of surplus sludge. Adv. Environ. Res. 2003, 7, 715-720. Cammarota, M. C.; Teixeira, G. A.; Freire, D. M. G. Enzymatic pre-hydrolysis and anaerobic degradation of wastewaters with high fat contents. Biotechnol. Lett. 2001, 23, 1591-1595. Jeganathan, J.; Nakhla, G.; Bassi, A. Hydrolytic pretreatment of oily wastewater by immobilized lipase. J. Hazard. Mater. 2007, 145, 127-135. Jordan, S. N.; Mullen, G. J. Enzymatic hydrolysis of organic waste materials in a solidliquid system. Waste Manage. 2007, 27, 1820-1828. Leal, M.; Freire, D. M. G.; Cammarota, M. C.; Sant'Anna, G. L. Effect of enzymatic hydrolysis on anaerobic treatment of dairy wastewater. Process Biochem. 2006, 41, 1173-1178. Parmar, N.; Singh, A.; Ward, O. P. Enzyme treatment to reduce solids and improve settling of sewage sludge. J. Ind. Microbiol. Biotechnol. 2001, 26, 383-386. Roman, H. J.; Burgess, J. E.; Pletschke, B. I. Enzyme treatment to decrease solids and improve digestion of primary sewage sludge. Afr. J. Biotechnol. 2006, 5, 963-967. Recktenwald, M.; Wawrzynczyk, J.; Dey, E. S.; Norrlow, O. Enhanced efficiency of industrial-scale anaerobic digestion by the addition of glycosidic enzymes. J. Environ. Sci. Health., Part A. 2008, 43, 1536 - 1540. Wawrzynczyk, J.; Recktenwald, M.; Norrlow, O.; Dey, E. S. Solubilisation of sludge by combined chemical and enzymatic treatment. Afr. J. Biotechnol. 2007, 6, 1994-1999. Valladao, A. B. G.; Freire, D. M. G.; Cammarota, M. C. Enzymatic pre-hydrolysis applied to the anaerobic treatment of effluents from poultry slaughterhouses. Int. Biodeterior. Biodegrad. 2007, 60, 219-225. Kim, H. J.; Kim, S. H.; Choi, Y. G.; Kim, G. D.; Chung, T. H. Effect of enzymatic pretreatment on acid fermentation of food waste. J. Chem. Technol. Biot. 2006, 81, 974980.

Water Production and Wastewater Treatment, Nova Science Publishers, Incorporated, 2010. ProQuest Ebook Central,

Copyright © 2010. Nova Science Publishers, Incorporated. All rights reserved.

26

Debora Nabarlatz, Frank Stüber, Josep Font et al.

[23] Zhang, J. F.; Gong, J. X.; Shao, G. Q.; Qin, J. J.; Gu, Z. Y. Biodegradability of diethylene glycol terephthalate and poly(ethylene terephthalate) fiber by crude enzymes extracted from activated sludge. J. Appl. Polym. Sci. 2006, 100, 3855-3859. [24] Gavala, H. N.; Yenal, U.; Ahring, B. K. Thermal and enzymatic pretreatment of sludge containing phthalate esters prior to mesophilic anaerobic digestion. Biotechnol. Bioeng. 2004, 85, 561-567. [25] APHA. Standard Methods for the Examination of Water and Wastewater. American Public Health Association. American Water Association: Washington, 1999. [26] Nabarlatz, D.; Vondrysova, J.; Jenicek, P.; Stüber, F.; Font, J.; Fortuny, A.; Fabregat, A.; Bengoa, C. Extraction of enzymes from activated sludge. In Waste Management and the Environment IV. WIT Transactions on Ecology and the Environment; M. Zamorano, et al.; Ed. WIT Press: Southampton, 2008; pp 249 - 257. [27] Nabarlatz, D.; Vondrysova, J.; Jenicek, P.; Stüber, F.; Font, J.; Fortuny, A.; Fabregat, A.; Bengoa, C. Extraction of protease and lipase from activated sludge by ultrasound and magnetic stirring disintegration. In 18th International Congress of Chemical and Process Engineering. Process Engineering Publisher: Prague, Czech Republic, 2008. [28] Voet, D.; Voet, J.; Pratt, C. Fundamentals of Biochemistry: Life at the molecular level. 2nd ed. Editorial Medica Panamericana: Madrid, Spain, 2006; pp 1130. [29] Goel, R.; Mino, T.; Satoh, H.; Matsuo, T. Enzyme activities under anaerobic and aerobic conditions inactivated sludge sequencing batch reactor. Water Res. 1998, 32, 2081-2088. [30] Boczar, B. A.; Forney, L. J.; Begley, W. M.; Larson, R. J.; Federle, T. W. Characterization and distribution of esterase activity in activated sludge. Water Res. 2001, 35, 4208-4216. [31] Schade, M.; Lemmer, H. Lipase activities in activated sludge and scum - Comparison of new and conventional techniques. Acta Hydroch. Hydrob. 2005, 33, 210-215. [32] Li, Y.; Chrost, R. J. Microbial enzymatic activities in aerobic activated sludge model reactors. Enzyme Microb. Technol. 2006, 39, 568-572.

Water Production and Wastewater Treatment, Nova Science Publishers, Incorporated, 2010. ProQuest Ebook Central,

In: Water Production and Wastewater Treatment Editors: B. Antizar-Ladislao and R. Sheikholeslami

ISBN 978-1-61728-503-5 © 2011 Nova Science Publishers, Inc.

Chapter 3

DGGE AND 16S RDNA SEQUENCING ANALYSIS OF BACTERIAL COMMUNITIES IN A MEMBRANE BIOREACTOR FOR THE REMOVAL OF PHENOL FROM OIL REFINERY WASTEWATER Fernanda R. Pinhatia, Aline F. Vieroa, Eduardo M. Del Aguilaa, Ana Paula R. Torresb, Joab T. Silvaa, and Vânia M. F. Paschoalina1* a

Copyright © 2010. Nova Science Publishers, Incorporated. All rights reserved.

Universidade Federal do Rio de Janeiro, Instituto de Química, Av. Athos da Silveira Ramos, 149 - Bloco A, sala 545, Cidade Universitária, 21941-909, Rio de Janeiro, RJ, Brazil b Petrobras-CENPES, Gerência de Biotecnologia e Tratamentos Ambientais, Av. Horácio de Macedo, 950 Cidade Universitária, 21941-915 Rio de Janeiro, RJ, Brazil

ABSTRACT The stability of the activated sludge bacterial community present in a submerged membrane bioreactor (SMBR) fed with petroleum refinery wastewater was investigated. The changes in the activated-sludge bacterial community during adaptation from low- to high-phenol loading were analyzed using a 16S rDNA-based technique. Amplicons from the V3 variable regions of bacterial 16S rDNA were analyzed by denaturing gradient gel electrophoresis (DGGE) and sequencing. The phenol removal efficiency of the SMBR bioreactor was 98% during acclimation (15 mg phenol L-1), high-organic loading (85 mg phenol L-1), and recovery (15 mg phenol L-1) steps. The cluster analysis of DGGE fingerprints showed higher similarity between the community structure of the bacterial population in the SMBR when the community structure of ‘native’ sludge was compared to bacterial communities during the high-phenol and recovery steps, with Cs of 76.7 and 63.4%, respectively. The fingerprint data combined with statistical tools showed that the bacterial community could adapt to the adverse environmental conditions during 1

*Corresponding author: e-mail: [email protected], tel +55 21 2562 7362; fax: +55 21 2562 7266.

Water Production and Wastewater Treatment, Nova Science Publishers, Incorporated, 2010. ProQuest Ebook Central,

28

Fernanda R. Pinhati, Aline F. Viero, Eduardo M. Del Aguila et al. wastewater treatment operations, sustaining the high quality of the effluents and maintaining stability of the SMBR in response to high-phenol-shock loadings. Seven prominent bands excised from DGGE fingerprints from the phenol-shock bands were successfully reamplified and sequenced, allowing the identification of predominant bacteria in the sludge. The phylogenetic analysis indicated that the isolates fell into two major lineages of the Bacteria domain: the Alpha and Gamma proteobacteria classes.

Copyright © 2010. Nova Science Publishers, Incorporated. All rights reserved.

INTRODUCTION Phenolic compounds are the major pollutants resulting from many industrial processes, including refining petroleum, synthesizing resins, and manufacturing explosives (Geng et al. 2006). Phenol is toxic to many biochemical functions and to marine life, even at low concentrations (Prieto et al. 2002). In human beings, phenols can cause liver and kidney damage and blood pressure drop (Barrios-Martinez (2006), and are toxic by ingestion, contact, or inhalation. Brazilian law specifies the concentration of 0.5 mg L-1 as the limit for discharge of phenol in effluents (Conama, 2005). Granular or biological activated-carbon filtration, ozonation, solvent extraction, and membrane-filtration processes are used to remove phenolic compounds from wastewater (Nuhoglu and Yalcin, 2005; Barrios-Martinez et al. 2006). However, the microbial-based process for phenol removal from industrial wastewater has generated significant interest because phenols and other aromatic compounds can be used as the sole carbon and energy source by many microorganisms. Biological treatment is a desirable alternative to traditional physical and chemical methods because of its low cost, reliable operational stability, and efficient destruction/reduction of pollution (Jiang et al. 2004). Although phenol is inhibitory at high concentrations, biological techniques are widely used for treatment of wastewater or soil containing moderate levels (5–500 mg L-1) of this compound (Barrios-Martinez et al. 2006). Biodegradation of phenol is often a prerequisite for the treatment of mixed pollutants, and phenol concentration can be accepted as an indicator of the removal efficiency of pollutant compounds from industrial wastewater, including refinery effluents (Margesin et al. 2005). Conventional wastewater treatment in oil refinery wastewater plants comprises several steps, generally consisting in oil/water separation, biological treatment and filtration/adsorption as an optional additional step. The process generally produces good results, although the activated sludge used in the biological treatment step is poorly suited to deal with large changes in flow rate and/or composition (Jiang et al. 2004). Activated sludge from aerobic wastewater-treatment plants is formed by a complex consortium of microorganisms, all of which are required to achieve the desired biological conversions. The microbial consortium from activated sludge has been exhaustively studied in order to understand the specific role of each microbial population in the biological processes and the effect of the microbial diversity on the overall performance (Amann et al. 1997; Liu et al. 2007; Schwartz et al. 2000). Membrane bioreactors (MBRs), which can be defined as integrating a biological degradation system of waste products with membrane filtration (Molina-Muñoz et al. 2007), has many advantages over conventional wastewater treatment technologies, including high quality of effluent, small footprint, less excess sludge, and ease of operation (Ueda and

Water Production and Wastewater Treatment, Nova Science Publishers, Incorporated, 2010. ProQuest Ebook Central,

Copyright © 2010. Nova Science Publishers, Incorporated. All rights reserved.

DGGE and 16S rDNA Sequencing Analysis of Bacterial Communities…

29

Horan, 2000). The use of MBRs is a very suitable technology for decentralized wastewater management and satellite wastewater-treatment systems (Lobos et al. 2005). The development of cost-effective manufacturing of membranes and increasingly stringent regulations for the discharge of effluents have accelerated the application of membrane technology in wastewater treatment and reuse (Choi et al. 2005). Many full-scale wastewater treatment plants using MBRs are already in operation, and the number of installations of MBRs is expected to increase. The water quality of MBR effluent is suitable for a variety of water-reuse applications, and is suitable, after disinfection, for a number of non-potable but otherwise unrestricted uses (Laera et al. 2005; Gobel et al. 2007). In the submerged MBR (SMBR) process, the membrane is immersed directly in the aeration tank. By applying low vacuum or by using the static head of the mixed liquor, effluent is driven through the membrane, leaving the solids behind. The advantages of the SMBR process for wastewater treatment include high sludge concentration, high quality of effluent, long contact time between the activated sludge and organic pollutants, and complete separation of the hydraulic retention time (HRT) and sludge retention time (SRT) (Chiemchaisri et al. 1992). Moreover, highly treated water in an SMBR is free from bacteria and has the potential for municipal and industrial reuse (Schneider and Tsutiya, 2001). Studies on the effects of shock loadings in MBR fed with industrial wastewater are few (Banerjee, 1997; González et al. 2001; Jou and Huang, 2003, Hsien and Lin, 2005), especially on oil refinery wastewater, which contains many volatile aromatic compounds, polycyclic aromatic hydrocarbons, and inorganic compounds. Because of the diversity and complexity of refinery wastewater, phenols have been accepted as a suitable compound to provide an indication of the performance of biodegradation (Barrios-Martinez, 2006). Viero et al. (2008) studied the effects of organic- shock loadings with respect to the organic matter, ammonianitrogen, and phenol removal efficiencies, and proved the ability of the SMBR to manage high-strength feed during long-term exposure, achieving high efficiencies in removing phenols and ammonia-nitrogen. However, the changes in the microbial community during shock loading have not been assessed. Most living bacteria have not yet been isolated and characterized (Torsvik et al. 1990; Giovannoni et al. 1991), because most of them cannot be cultured by standard techniques (Liu et al. 2007). For example, the culturable fraction of the bacteria present in wastewatertreatment reactors was estimated to be typically around 15 – 20% (LaPara et al. (2002), and the microbial communities found in refinery activated-sludge systems are still uncharacterized. Recently, several molecular techniques have been developed to study microbial communities. Methods based on direct PCR amplification of 16S rRNA gene and analysis of the amplified fragment by temperature-gradient gel electrophoresis (TGGE) or denaturing-gradient gel electrophoresis (DGGE) have been frequently used to examine the microbial diversity of environmental samples and to monitor changes in microbial communities (Amann et al. 1995). DGGE methods are relatively simple and produce results in relatively short time (Miura et al. 2007). They are based on the separation of the amplified fragments according to their nucleotide sequences, which are specie-specific. DGGE has provided important information about the diversity of microorganisms in natural and engineered habitats, including those microbial species previously unknown because of the limitations of culture-based approaches (Wagner and Loy, 2002; Cortés-Lorenzo et al. 2006). This technique has been frequently used to study the ecology of biological processes in MBRbased wastewater treatment plants, and to compare communities from different reactors or

Water Production and Wastewater Treatment, Nova Science Publishers, Incorporated, 2010. ProQuest Ebook Central,

30

Fernanda R. Pinhati, Aline F. Viero, Eduardo M. Del Aguila et al.

from the same reactor operated under different conditions (Witzig et al. 2002; Stamper et al. 2003; Miura et al. 2007; Molina-Muñoz et al. 2007). In this study, we investigated the changes over time in the bacterial community present in the activated sludge from an SMBR fed with oil refinery effluent, during long-term phenolshock loadings, by comparing the DGGE fingerprints of the V3 hyper-variable region of the 16S rDNA produced from total DNA extracted periodically from the activated sludge present in the bioreactor.

MATERIAL AND METHODS SMBR Operation

Copyright © 2010. Nova Science Publishers, Incorporated. All rights reserved.

The reactor construction and operation have been described previously (Viero et al. 2008). In brief, a mini-reactor consisting of a cylindrical acrylic tank with a working volume of 4.4 L was used. The polyetheremide membranes (average pore diameter of 0.15 ± 0.09 µm and total surface area 2.78×10−2m2) were produced by wet spinning, using the phaseinversion process. The bioreactor was operated at 25°C under a hydraulic retention time (HRT) of 10 h, air flow rate 2.5 L min−1, filtration time 5 min, backwashing time 6.25 min, air backwashing pressure 3.0 bars, and liquid permeate flux 15–17 L m-2 h-1 bar-1. The reactor was fed with a mixture of effluents collected from a petroleum refinery, consisting of an oily stream (OS) from the oil desalting process mixed with acid wastewater, and oily water drained from crude-oil storage tanks. A phenolic wastewater (PW) drained from the bottoms of cracked-gasoline tanks was added to increase the content of recalcitrant compounds content in the feed. The NH4+-N content in the PW and OS was 562.5 mg L-1 and 110 mg L-1, respectively.

Activated Sludge and Oil-Refinery Wastewater The activated sludge (10 g L−1 of total suspended solids after sedimentation) was collected in a Brazilian oil refinery wastewater-treatment plant. The sludge was acclimated for 33 days in the SMBR processing the regular oily wastewater. During acclimation and recovery, the SMBR operated with the wastewater flow rates regularly used in the refinery (15 mg L−1 phenol). After acclimation, the SMBR was fed with a mixture of oil wastewater (OS) and phenolic wastewater, showing mean CN-1 concentrations of 0.34 mg L-1; Cl-1, 1084 mg L-1; and NH4+-N concentrations of 110 mg L-1; S-2 of 2.23 mg L-1, respectively. At the stage of high organic-loading operation, the phenol concentration was raised to 85 mg L−1.

Analytical Methods Samples of effluent were collected in an oil refinery in Brazil. The initial effluent and the effluent samples collected periodically from the SMBR reactor were assayed for chemical oxygen demand (COD) and concentrations of phenol, CN−1, Cl−1, NH4+-N and S−2 (Standard

Water Production and Wastewater Treatment, Nova Science Publishers, Incorporated, 2010. ProQuest Ebook Central,

DGGE and 16S rDNA Sequencing Analysis of Bacterial Communities…

31

Methods, 1998), using the Water Testing Kits (Chemetrics Inc) as recommended by the manufacturer.

Sampling of Activated Sludge After acclimation (day 34), a sample of the activated sludge present in the SMBR was collected to characterize the ‘native’ sludge. During the high-loading-rate shock (days 35 to 63), activated sludge samples were collected each day, and during the last phase (days 64 to 90), when the loading-rate feed in the process was similar to the acclimation period, activated-sludge samples were taken every two days. Samples of activated sludge were collected by centrifugation at 2,000 g for 15 min, and the pellets were stored at -20 oC prior to DNA extraction.

Copyright © 2010. Nova Science Publishers, Incorporated. All rights reserved.

DNA Extraction from Sludge Samples DNA was extracted from activated-sludge samples according to the method of Melvin & Hobson (1994), with some modifications. Each sludge pellet (0.5 g wet weight) was suspended in 1mL of TESC buffer (10mM Tris, 0.1mM EDTA and 0.1mM NaCl), pH 8.3, and shaken vigorously. After incubation for 2 h at –80ºC, 5 µL of DMSO (dimetyl sulfoxide) were added, and the suspension was again incubated at room temperature for 1 min. Subsequently, 500 µL of 5M guanidine thiocyanate and 500 µL of phenol:chloroform:isoamyl alcohol (25:24:1 v/v) were added. After centrifugation, the aqueous phase was transferred to a new tube and 30 µL of 3M sodium acetate was added. Total nucleic acids were precipitated with 2.5 volumes of 100% ethanol (-20ºC), dried, suspended in sterile double-distilled water, and then stored at –20ºC.

PCR Amplification The V3 hypervariable region of the 16S rRNA gene, corresponding to nucleotide positions from 968 to 1401 (Escherichia coli), was amplified using the universal bacterial primers 968f (5’-AACGCGAAGAACCTTAC-3’) containing a 40-bp GC clamp (5’CGCCCGCCGCGCGCGGCGGGCGGGGCGGGGGCACGGGGGG-3’) added to its 5’-end, and 1401r (5’-CGGTGTGTACAAGACCC-3’) described by Nübel et al. (1996). The PCR mixture comprised 5 µl of the DNA preparation extracted from sludge samples, 50 pmol of each universal primer, 5 µl of 10X PCR buffer (Invitrogen), 0.1µM MgCl2, 0.2mM of each deoxynucleoside triphosphates (Invitrogen), and sterile ultrapure water, to a final volume of 50 µl. Negative controls consisting of sterile ultrapure water instead of the sample were included in each batch of samples. PCR was performed in a Perkin Elmer Gene Amp® PCR System 2400, with an initial denaturation step of 94 °C for 7 min, followed by 25 cycles of a denaturation step at 94ºC for 1 min, a primer annealing step at 60ºC for 1 min, an extension step at 72ºC for 1 min, and a final step of 72ºC for 7 min. Prior to the DGGE analysis, the presence of the PCR product of 433 bp was confirmed by electrophoresis in a

Water Production and Wastewater Treatment, Nova Science Publishers, Incorporated, 2010. ProQuest Ebook Central,

32

Fernanda R. Pinhati, Aline F. Viero, Eduardo M. Del Aguila et al.

1.2% agarose gel stained for 15 min with 0.5 µg/ml ethidium bromide. A 100-bp DNA ladder digest (Invitrogen) served as the molecular size standard.

DGGE Assays DGGE banding profiles of the V3 region of the 16S rDNA amplified by PCR with the 968f-GC/1401r primers were obtained using the Dcode Universal Mutation Detection System (Bio-Rad Dcode, Hercules, CA, USA) at 50 V and 60ºC for 14 h in 0.5X TAE buffer (20 mM Tris-acetate 10 mM sodium acetate, 0.5 mM disodium EDTA, pH 7.4). The PCR products (30 µl) were loaded on 6% (w/v) polyacrylamide gels containing a linear gradient of the denaturants urea and formamide (45 to 65%), increasing from the top to the bottom of the gel [100% denaturant corresponding to 7 M urea and 45% (v/v) formamide]. Two gels were run to accommodate all the samples. A sample showing multiple bands in earlier experiments was loaded in the first and last slots of each gel to facilitate alignment and comparison between gels. After electrophoresis, gels were stained with SYBR green I nucleic acid gel stain (1:10.000 dilution; Molecular Probes, USA) for 60 min and observed/documented under UV light with the MiniBisPro System (BioAmerica Inc).

Copyright © 2010. Nova Science Publishers, Incorporated. All rights reserved.

Analysis of DGGE Fingerprints Cluster analysis was used to estimate the relatedness of the DGGE profiles representing the activated-sludge bacterial community in the SMBR. Band patterns generated by DGGE were normalized, compared, and clustered using the Gel Compar II v. 5.0 software (Applied Maths, Belgium). Bands were automatically detected and matched, with additional manual fine tuning of the band designations. Dendrograms relating band pattern similarities were automatically calculated with the Dice coefficient and UPGMA algorithms (Unweighted pair group method with arithmetic mean). The significance of UPGMA clustering was estimated by calculation of the cophenetic correlation coefficients.

Sequencing of DGGE Bands After DGGE, bands of interest were excised from the gel and the DNA from each band was extracted using the QIAEX II GEL Extraction Kit (QIAGEN, Hilden, Germany). Subsequently, an aliquot (3 µl) was used to amplify the DGGE bands using the primers 968F (without the GC clamp) and 1401R, under the same amplification conditions. The amplified bands were purified using the Wizard SV Gel and PCR Clean-Up System (Promega, WI, USA). The DNAs recovered were used for automated sequencing in an ABI PRISM 3100 Avant Genetic Analyzer. DNA sequences were used to classify bacteria to the genus level by using the biocomputing tool Sequence-Match from the Ribosomal Database Project (http://rdp.cme.msu.edu/seqmatch).

Water Production and Wastewater Treatment, Nova Science Publishers, Incorporated, 2010. ProQuest Ebook Central,

DGGE and 16S rDNA Sequencing Analysis of Bacterial Communities…

33

RESULTS AND DISCUSSION Evaluating the Efficiency of SMBR

Copyright © 2010. Nova Science Publishers, Incorporated. All rights reserved.

The SMBR was operated for 90 days, fed with refinery wastewater (RW) showing mean values of chemical oxygen demand (COD) and phenol of 616 mg L-1 and 15 mg L-1, respectively. The RW was additionally characterized: CN−1 (0.34 mg L−1), Cl−1 (1084 mg L−1), NH4 +-N (110 mg L−1) and S−2 (2.23 mg L−1). The stock phenol wastewater (PW) showed mean values of COD, phenols and NH4+-N of 55.754 mg L-1, 628 mg L-1 and 562.5 mg L-1, respectively. Both the PW and RW were generated in the same oil refinery where the activated sludge was collected. The average phenol removal by the SMBR was always higher than 98% (Figure 1A), and reached values greater than 99.3% when the phenol concentration in the feed was less than 15 mg L−1, as in the recovery phase. Notably, phenol removal efficiency remained high even during the high-organic-loading operation phase, when the phenol concentration was raised to 85 mg L−1 in the feed, but the phenol in the SMBR permeate was raised to the still-low concentration of 0.35 mg L-1 (Figure 1B), within the specifications laid down by Brazilian legislation for the waste effluent (Conama, 2005), which established that concentrations of phenol in effluents must be less than 0.5 mg L-1. The average removal efficiency of phenol obtained in this study (98%) was similar or superior to those reported in the literature.

Figure 1. Reactor performance. A) Phenol concentration in the feed and phenol removal efficiency. B) Phenol concentration in the permeates.

This observed efficiency is even greater if we consider that the influent contained significant amounts of NH4 +-N, which is potentially toxic to sludge microbes and could reduce the efficiency of removal of phenols and organic matter. The removal efficiencies were high, considering that the hydraulic retention time employed was 10 h. Vázquez et al. (2006) described a phenol biodegradation process for coke wastewater treatment using a laboratory-scale activated sludge plant composed of a 20 L volume aerobic reactor, which

Water Production and Wastewater Treatment, Nova Science Publishers, Incorporated, 2010. ProQuest Ebook Central,

34

Fernanda R. Pinhati, Aline F. Viero, Eduardo M. Del Aguila et al.

achieved 96% efficiency with a hydraulic retention time (HRT) of 15 h. A similar membrane bioreactor performance was also described by Barrios-Martinez et al. (2006) who reported a phenol removal efficiency of 100% during the treatment of a synthetic wastewater containing phenols at concentrations of 1000mg L−1 using adapted biomass. The ability to remove large amounts of phenol, together with the good retention of other organic substances, demonstrated the proper performance of the SMBR. This method should be viewed as an alternative to the traditional physical and chemical methods for phenol removal, generally considered to be costly and dangerous to handle (H2O2, O3).

Copyright © 2010. Nova Science Publishers, Incorporated. All rights reserved.

Communities Structure Analysis The influence of the phenol-shock loading on the SMBR bacterial community was studied by DGGE. Bacterial community changes were detected by isolation of total DNA from activated sludge from SMBR during a 93-day time course of sampling. Twenty-four samples were analyzed: 01 collected at day 34, after acclimatization of the sludge, 12 samples during the phenol-shock loading (days 35 to 63), and 11 samples during the recovery period (days 64 to 93). DGGE banding profiles of the PCR-amplified V3 region of the 16S rDNA were produced from bacterial community DNA extracted from the above activated-sludge samples. Experiments designed to establish the repeatability of DNA extraction and PCR amplification found a 90% to 95% similarity in triplicate independent samples, as determined by analysis of the DGGE patterns (results not shown). The analysis of complex microbial communities by DGGE technique is limited by the difficulty to extract and amplify by PCR the DNA of unknown communities (Gelsomino et al. 1999) and the possibility of formation of chimeric PCR products (Wang and Wang, 1997), reducing the variability of microorganisms contained in each sample. However, the DGGE fingerprints obtained from the sludge samples collected during the 93 days of SMBR continuous operation detected a great genetic variety of the microorganisms (Figure 2A). The community structure was remarkably dynamic. None of the bands was universally present in the samples, and 24 band classes were detected in over 72% of them, compared to 6 band classes being represented in 18% or less of the samples. DGGE fingerprints were of medium complexity, as 11–22 bands per lane were recorded, distributed in a total of 33 different band classes. This number is within the range reported by previous studies on sludge from MBRs (Stamper et al. 2003; van der Gaast et al. 2006; Molina-Muñoz et al. 2007), but bacterial communities found in aerobic wastewater-treatment systems based on conventional activated sludge or submerged filter biofilms, are usually more complex (Wagner and Loy, 2002; Gómez-Villalba et al. 2006). Bacterial diversity in MBRs has been also found to be strongly influenced by bioreactor volume (van der Gaast et al. 2006), and thus the small volume pilotscale bioreactor used in this study is expected to naturally support lower levels of diversity than full-scale MBRs.

Water Production and Wastewater Treatment, Nova Science Publishers, Incorporated, 2010. ProQuest Ebook Central,

Copyright © 2010. Nova Science Publishers, Incorporated. All rights reserved.

DGGE and 16S rDNA Sequencing Analysis of Bacterial Communities…

35

Figure 2. DGGE banding profiles of the V3 region of 16S rDNA produced from bacterial community DNA extracted from different samples of the activated sludge present in MBR during shock loading. A) DGGE was performed as described in Material and Methods, and stained with SYBR Green I. Acclimated activated sludge (lane 34), high phenol-shocked sludge (lanes 35 to 63), and recovered sludge (lanes 64 to 91). Arrows indicate the bands that were sequenced. B) Clustering of the DGGE profiles obtained with universal V3 primer using UPGMA method and Dice’s similarity coefficient computed using Gel Compar II v. 5.0 (Applied Maths, Belgium). The scale bar indicates the percentage of similarity.

The results suggested that the bacterial community structure was dynamic and able to adapt to environmental changes. Several researchers have reported that the bacterial community structure of laboratory-scale activated-sludge reactors seeded with sludge from

Water Production and Wastewater Treatment, Nova Science Publishers, Incorporated, 2010. ProQuest Ebook Central,

Copyright © 2010. Nova Science Publishers, Incorporated. All rights reserved.

36

Fernanda R. Pinhati, Aline F. Viero, Eduardo M. Del Aguila et al.

domestic wastewater treatment plants was not static, but constantly changed. Interestingly, this dynamic behavior of the bacterial community of laboratory-scale bioreactors is not observed in full-scale biological treatment plants (Godon et al. 1997; Eichner et al. 1999; Boon et al. 2000; Forney et al. 2001). LaPara and colleagues (2002), examining the bacterial community structure of seven fullscale biological plants treating pharmaceutical wastewater, showed that the community was stable under normal operating conditions. A stable community structure in full-scale biological treatment plants compared to laboratory-scale activated-sludge reactors could be explained by the equilibrium model of island biogeography, which seeks to establish and explain the factors affecting the species richness of a particular isolated community (Curtis et al. 2003). A consortium of bacterial species is required for achieving the desired biological conversions, and the performance of these reactors largely depends on the bacterial diversity present (Saikaly and Oerther 2004). This dynamic behavior in the bacterial community structure in laboratory-scale activatedsludge reactors could be attributed to several biotic and abiotic factors, such as resource competition (Huisman and Weissing, 1999; 2001), predation, and new selective pressure imposed on domestic sludge. It is recognized in ecology that competition for three or more growth-limiting resources may generate oscillations and chaotic fluctuations in species abundances (Huisman and Weissing, 2002). Recently, Saikaly and Oerther (2004) developed an ecology-based mathematical model to describe the mechanism behind these chaotic bacterial community structures. Although the DGGE technique alone does not allow for the complete characterization of complex communities such as that inhabiting wastewater-treatment sludge, it provides an efficient tool to monitor the dynamics of their species composition, as influenced by external parameters (Lyautey et al. 2005). The hierarchical grouping of bacterial communities carried out using UPGMA method and Dice’s similarity coefficient computed using Gel Compar II v. 5.0 (Applied Maths, Belgium) (Fig. 2B), allowed the identification of four clusters that reflect changes in bacterial community following the change in concentration of phenol in effluent fed into the reactor. The community present during the operation at a high phenol loading (day 35 to 63, stage 2) was comprised by two clusters. The first one included the “native” sludge (day 34) and the sludge community from samples collected at days 35, 36, 37 and 40. The second cluster included the samples collected at days 41, 42, 43, 44, 50 and 55. In contrast, the sludge bacterial community from samples collected in the recovery phase (days 64 to 90) was grouped into one cluster that included the profile of the sample collected at day 63 (last day of operation of the high-phenol-loading regimen), and formed three sub-clusters. The fourth cluster was constituted from profiles of the samples collected at day 58 (high-phenol loading) and 90 (recovery), sharing 77.9% similarity, and clustering away from the rest of the samples. These profiles are characterized by a low number of band classes (17 and 9, respectively) and may represent stages of decrease in microbial diversity under changing environmental conditions. Most of the fingerprints from samples collected during the phenol-shock loading (days 35 to 63) and recovery (days 64 to 90) stages were grouped together, and showed a 54.8% overall similarity. The similarity between the profiles from the high-phenol loading and recovery stages was 63.4% and 76.7%, respectively. As can be observed, genetic diversity decreased during the course of the treatment because, probably, some of the original

Water Production and Wastewater Treatment, Nova Science Publishers, Incorporated, 2010. ProQuest Ebook Central,

DGGE and 16S rDNA Sequencing Analysis of Bacterial Communities…

37

microorganisms present in the acclimated sludge were selected during the high-phenolloading step. The plasticity of the microbial community ensured the efficiency of phenol degradation and the successful performance of the SMBR (Fernández et al. 1999). Individual persistent species responsible for these bands would be likely candidates for incorporation into a seed culture and improve bioreactor performance, particularly during startup of wastewater treatment (Stamper et al. 2003).

Copyright © 2010. Nova Science Publishers, Incorporated. All rights reserved.

Communities Shifts During Phenol-Shock Loadings Dice’s similarity coefficient (Cs) computed using Gel Compar II v. 5.0 (Applied Maths, Belgium) was used to evaluate the similarity of the DGGE fingerprints from sludge samples collected during the SMBR operation. The Cs for the “native” sludge (day 34) in comparison to sludge collected at day 63 or day 90 was 49.8% and 88.2%, respectively. A decrease in similarity was observed between the acclimated sludge (34 day) and that collected at the end of the phenol-shock loading (day 63), however, the Cs increased when the acclimated sludge was compared to the recovery loading, on the last operating day (day 90). Dynamic changes in SMBR communities may be related to the toxicity of phenol to the sludge microorganisms. In petrochemical wastewater treatment plants, phenolic compounds, even at low concentrations, can inhibit microbial growth. Uygur and Kargi, (2004) reported variable inhibition of nutrient removal from wastewater by a sequencing bath reactor (SBR) at different phenol concentrations. Phenol toxicity to microorganisms may be related to the reduction of observed bands in the DGGE samples taken during the high phenol-loading (Fig. 2B, days 41 to 55) or even during the recovery loading (Fig. 2B, days 64 to 85), which may select some resistant species during these periods. However, the structure of bacterial communities in the batch bioreactor was virtually stable throughout the process, according to the analysis of the indexes of similarity comparing the bacterial community present in the SMBR at the high phenol load and recovery load with the ‘native’ sludge (76.7% and 63.4%) and the large number of bands that remained in the DGGE fingerprints throughout the SMBR operation. Activated sludge, similarly to the other diverse ecosystems such as sediments and soils, has DGGE banding patterns that are very complex to interpret. Moreover, analyses of similarity are necessary to examine and compare the DGGE fingerprints generated. However, the bands in a DGGE fingerprint do not necessarily give an accurate picture of the number and abundance of the corresponding species within the microbial community. One organism may produce more than one DGGE band because of multiple, heterogeneous rRNA operons (Cilia et al. 1996). On the other hand, partial 16S rDNA sequences amplification does not always allow discrimination between species, since one DGGE band may represent several species with identical partial 16S rDNA sequences (Naeem and Li, 1997). In addition, in a mixture of target rDNAs with different concentrations, the less-abundant sequences cannot be sufficiently amplified to be observed. For this reason, the banding pattern reflects only the most abundant rDNA types in the microbial community (Huisman and Weissing, 1999). Therefore, the similarity index calculated from the DGGE fingerprints of amplified 16S rDNA sequences must be interpreted as an indication and not an absolute measure of the degree of complexity in a bacterial community (Curtis and Sloan, 2004). In our study, the DGGE fingerprints were related to samples taken from the same “native sludge” exposed to

Water Production and Wastewater Treatment, Nova Science Publishers, Incorporated, 2010. ProQuest Ebook Central,

38

Fernanda R. Pinhati, Aline F. Viero, Eduardo M. Del Aguila et al.

different concentrations of phenol, and not from environmental samples collected in diverse ecosystems. The equal band intensities in samples collected during SMBR operation at the same feed condition indicated the absence of preferential amplification. Furthermore, there is no evidence of chimera or heteroduplex formation, since no additional bands were detected in the DGGE profiles. The PCR-DGGE results presented herein indicate that functionally stable wastewater treatment bioreactors also have stable microbial community structures and are capable of adapting in response to perturbations. We believe that PCR-DGGE combined with another nucleic acid-based technique (e.g., ARDRA and in situ hybridization) can be used in future research to develop a mechanistic understanding of the relationships between reactor operational strategies, microbial community structure, and reactor performance.

Phylogenetic Analysis

Copyright © 2010. Nova Science Publishers, Incorporated. All rights reserved.

A total of seven bands from the shock-loading period (day 43 – Fig. 2A, indicated by arrows) were isolated from the DGGE gels, reamplified, and sequenced. The resulting nucleotide sequences were compared to those filed in the RPD database. Table 1 shows the identified sequences, which are mostly related to the Alpha-proteobacteria (bands 1, 3, 4 and 5) and Gamma-proteobacteria (bands 2, 6 and 7). Sequencing of DGGE bands corresponding to the dominant population in the sludge community profiles showed the prevalence of genus Acetobacter in all the experiments, in spite of the different operation conditions applied. Table 1. Seven prominent bands present in DGGE profiles of the V3 region of the 16S rDNA from activated sludge present in MBR during shock loading (Fig. 2A, indicated by arrows) were excised from the gel, and the isolated DNA was amplified by PCR using primers 968F (without the GC clamp) and 1401R, and sequenced in an ABI PRISM 3100 Avant Genetic Analyzer. Nucleotide sequences were used to identify the putative donator by using SeqMatch from the Ribosomal Database Project (http://rdp.cme.msu.edu/seqmatch). Accession number is from GenBank Band

Phylum

Class

Order

Family

Genus

1

ProteoBacteria ProteoBacteria ProteoBacteria ProteoBacteria ProteoBacteria ProteoBacteria Proteobacteria

Alphaproteobacteria Gammaproteobacteria Alphaproteobacteria Alphaproteobacteria Alphaproteobacteria Gammaproteobacteria Gammaproteobacteria

Rhodospirillales

Acetobacteraceae

Acetobacter

Accession number AP011121

Chromatiales

Halothiobacillaceae

Thiofaba

DQ415810

Rhodospirillales

Acetobacteraceae

Acetobacter

AP011163

Rhodobacterales

Acetobacteraceae

Acetobacter

AP011121

Rhodospirillales

Acetobacteraceae

Acetobacter

AP011135

Xanthomonadales

Xanthomonadaceae

Aquimonas

GQ354936

Xanthomonadales

Xanthomonadaceae

Thermomonas

GQ389149

2 3 4 5 6 7

Water Production and Wastewater Treatment, Nova Science Publishers, Incorporated, 2010. ProQuest Ebook Central,

DGGE and 16S rDNA Sequencing Analysis of Bacterial Communities…

39

Geng et al. (2006) previously described the genus Acinetobacter as a novel phenoldegrading bacterium isolated from an activated sludge from an industrial wastewater treatment plant. The authors used a culture-enrichment technique. Jiang et al. (2004) used both culture-based and culture-independent technique to investigate the bacterial diversity cultivated in a sequencing batch reactor. The phylogenetic analysis indicated that the isolates felt into three major lineages of the Bacteria domain: the Beta and Gamma proteobacteria, and gram-positive high- G+C bacteria. In the present study, we showed that in the dominant populations in the sludge community profiles, there was a prevalence of Alpha and Gamma proteobacteria. Although the Alpha proteobacteria comprised the majority, the presence of Gamma proteobacteria agrees with the previous work of Bramucci et al. (2003), who also found this class using methodologies entirely based on the culture approach. In our study, the bacteria could not be unambiguously identified to species level, suggesting that most of the bacteria in our SMBR sludge have not been cultured or identified.

Copyright © 2010. Nova Science Publishers, Incorporated. All rights reserved.

CONCLUSIONS This study demonstrated the performance of a membrane bioreactor (SMBR) using previously activated sludge, for the degradation of an industrial effluent from an oil refinery, containing high phenol concentration (85 mg L-1). The good performance of the SMBR was accompanied by changes in the structure of bacterial communities present in the sludge. The use of PCR-DGGE allowed monitoring of these changes in the bacterial communities. Dominant members of the bacterial community in the SMBR were identified as belonging to the Alpha and Gamma proteobacteria. The integration of the fingerprint data with statistical tools showed that the bacterial community could adapt to the adverse environmental conditions during operation wastewater-treatment operation, sustained high effluent quality functionally keeping the reactor wastewater treatment stable in response to changing characteristics of wastewater effluent. The finding that operation and environmental parameters mostly influenced the evolution of the structure of the microbial populations in the SMBR, and the impact of these community features on bioreactor performance and quality of the sludge, contributes to the understanding of the community-function relationships, and will aid in improving the performance of the bioreactors.

REFERENCES Amann, R. I.; Ludwig, W.; Schkeifer K. H. (1995). Phylogenetic identification and in situ detection of individual microbial cells without cultivation. Microbiology, 59, 143-169. Amann, R.; Glockner, F. O.; Neef, A. (1997). Modern methods in subsurface microbiology: in situ identification of microorganisms with nucleic acid probes. FEMS Microbiology Reviews, 20, 191-200. Banerjee, G. (1997). Treatment of phenolic wastewater in RBD reactor, Water Research, 31, 705-714.

Water Production and Wastewater Treatment, Nova Science Publishers, Incorporated, 2010. ProQuest Ebook Central,

Copyright © 2010. Nova Science Publishers, Incorporated. All rights reserved.

40

Fernanda R. Pinhati, Aline F. Viero, Eduardo M. Del Aguila et al.

Barrios-Martinez, A.; Barbot, E.; Marrot, B.; Moulin, P.; Roche, N. (2006). Degradation of synthetic phenol-containing wastewater by MBR. Journal of Membrane Science, 281, 288-296. Bramucci, M.; Kane, H.; Chen, M.; Nagarajan, V. (2003). Bacterial diversity in an industrial wastewater bioreactors. Applied Microbiology Biotechnology, 62, 594-600. Boon, N. J.; Goris, P.; De Vos, W.; Verstraete, K.; Top, E. M. ( 2000). Bioaugmentation of activated sludge by an indigenous 3-chloroanilinedegrading Comamonas testosteroni strain. Applied Environmental Microbiology, 66, 2906-2913. Cilia, V.; Lafay, B.; Christen, R. (1996). Sequence heterogeneities among 16S ribosomal RNA sequences, and their effect on phylogenetic analyses at the species level. Molecular Biology Evolution. 13, 451-461. Cortés-Lorenzo, C.; Molina-Muñoz, M.L.; Gómez-Lillalba, B.; Vilchez, R.; Ramos, A.; Rodelas, B. (2006). Analysis of community compostion of biofilms in a submerged filter system for the removal of ammonia and phenol from industrial wastewater. Biochemical Society Transactions, 334,165-168. Choi, H.; Kim, H.-S.; Yeom, L.-C.; Dionysiou, D. D. (2005). Pilot plant study of an ultrafiltration membrane system for drinking water treatment operated in the feed-andbleed mode, Desalination, 172, 281-291. Chiemchaisri, C.; Wong, Y. K.; Urase, T.; Yamamoto, K. (1992). Organic Stabilization and Nitrogen Removal in Membrane Separation Bioreactor for Domestic Wastewater Treatment. Water Science and Technology, 25, 231-240. CONAMA, (2005). Resolução 357, Diário Oficial da União, 53, 58-63. Curtis, T. P.; Head, I. M.; Graham, D. W. (2003). Theoretical Ecology for engineering biology. Environmental Science and Technology, 37, 64-70. Curtis, T. P. and Sloan W. T. (2004). Prokaryotic diversity and its limits: microbial community structure in nature and implications for microbial ecology. Current Opinion in Microbiology, 7, 221-226. Eichner, C. A.; Erb, R. W.; Timmis, K. N.; Wagner-Dobler, I. (1999). Thermal gradient gel electrophoresis analysis of bioprotection from pollutant shocks in the activated sludge microbial community. Applied Environmental Microbiology, 65, 102-109. Fernández, A.; Huang, S.; Seston, S.; Xing, J.; Hickey, R.; Criddle, C.; Tiedje, J. (1999). How stable is stable? Function versus community composition. Applied Environmental Microbiology, 65, 3697-3704. Forney, L. J.; Liu, W. T.; Guckert, J. B.; Kumagai, Y.; Namkung, E.; Nishihara, T.; Larson, R. J. (2001). Structure of microbial communities in activated sludge: Potential implications for assessing the biodegradability of chemicals. Ecotoxicology and Environmental Safety, 49, 40-53. Gelsomino, A.; Keijzer-Wolters, A. C.; Cacco, G.; van Elsas, J. D. (1999). Assessment of bacterial community structure in soil by polymerase chain reaction and denaturing gradient gel electrophoresis. Journal of Microbiological Methods, 38, 1-15. Geng, A.; Soh, A. E. W.; Lim, C. J.; Loke, L. C. T. (2006). Isolation and characterization of a phenol-degrading bacterium from an industrial activated sludge. Applied Microbiology Biotechnology, 71, 728-735. Giovannoni, S. J.; Britschgi, T. B.; Moyer, C. L.; Field, K. G. (1991). Genetic diversity in Sargasso Sea bacterioplankton, Nature, 345, 60-63.

Water Production and Wastewater Treatment, Nova Science Publishers, Incorporated, 2010. ProQuest Ebook Central,

Copyright © 2010. Nova Science Publishers, Incorporated. All rights reserved.

DGGE and 16S rDNA Sequencing Analysis of Bacterial Communities…

41

Gobel, A.; McArdell, C. S.; Joss, A.; Siegrist, H.; Giger, W. (2007). Fate of sulfonamides, macrolides and trimethoprim in different wastewater treatment technologies. Science Total Environment, 372, 361-371. Godon, J. J.; Zumstein, E.; Dabert, P.; Habouzit, F.; Moletta, R. (1997). Microbial 16S rDNA diversity in an anaerobic digester. Water Science Technology, 36, 49-55. Gómez-Vilalba, B.; Calvo, C.; Vilchez R.; González-López J.; Rodelas, B. (2006). TGGE analysis of the diversity of ammonia-oxidizing and denitrifying bacteria in submerged filter biofilms for the treatment of urban wastewater. Applied Microbiology Biotechnology, 72, 393-400. González, G.; Herrera, G.; García, M. T.; Peña, M. (2001). Biodegradation of phenolic industrial wastewater in a fluidized bed bioreactor with immobilized cells of Pseudomonas putida, Bioresource Technology, 80, 137-142. Hsien, T.-Y. and Lin, Y.-H. (2005). Biodegradation of phenolic wastewater in a fixed biofilm reactor, Biochemical Engineering Journal. 27, 95-103. Huisman, J. and Weissing, F. J. (1999). Biodiversity of plankton by species oscillations and chaos. Nature, 402, 407-410. Huisman, J. and Weissing, F. J. (2001). Biological conditions for oscillations and chaos generated by multispecies competition. Ecology, 82, 2682-2695. Huisman, J. and Weissing, F. J. (2002). Oscillations and chaos generated by competition for interactively essential resources. Evolutionary Ecology Research, 17, 175-181. Jiang, H-L.; Tay, J-H.; Maszenan, A. M.; Tay, S. T. L. (2004). Bacterial diversity and function of aerobic granules engineered in a sequencing batch reactor for phenol degradation. Applied Environment Microbiology, 70, 6767-6775. Jou, C.-J. G. and Huang, G.-C. (2003). A pilot study for oil refinery wastewater using a fixedfilm bioreactor. Advances in Environmental Research, 7, 463-469. Laera, G.; Pollice, A.; Saturno, D.; Giordano, C.; Lopez, A. (2005). Zero net growth in a membrane bioreactor with complete sludge retention. Water Research, 39, 5241-5249. LaPara, T. M.; Nakatsy, C. H.; Pantea, L. M.; Allegan, J. E. (2002). Stability of the bacterial communities supported by a seven-stage biological process treating pharmaceutical wastewater as revealed by PCR-DGGE. Water Research, 36, 638-646. Liu, X.-C.; Yang, M.; Zhang, Y. (2007). Microbial community comparison of different biological processes for treating the same sewage. Word Journal Microbiology Biotechnology, 23, 135-143. Lobos, J.; Wisniewshi, C.; Heran, M.; Grasmick, A. (2005). Effects of starvation conditions on biomass behaviour for minimization of sludge production in membrance bioreactors. Water Science Technology, 51, 35-44. Lyautey, E.; Lacoste, B.; Ten-Hage, L.; Rols, J-L.; Garabetian, F. (2005). Analysis of bacterial diversity in river biofilms using 16S rDNA PCR-DGGE: methodological settings and fingerprints interpretation. Water Research, 39, 380-388. Margesin, R.; Fonteyne, P. A.; Redl, B. (2005). Low-temperature biodegradation of high amounts of phenol by Rhodococcus spp. and basidiomycetous yeasts, Research in Microbiology, 156, 68-75. Melvin, W. T. and Hobson, P. N. (1994). Identification of anaerobic digester bacteria using a polymerase chain reaction method. Bioresource Technology, 47, 73-80.

Water Production and Wastewater Treatment, Nova Science Publishers, Incorporated, 2010. ProQuest Ebook Central,

Copyright © 2010. Nova Science Publishers, Incorporated. All rights reserved.

42

Fernanda R. Pinhati, Aline F. Viero, Eduardo M. Del Aguila et al.

Miura, Y.; Hiraiwa, M. N.; Ito, T.; Itonaga, T.; Watanabe, Y.; Okabe, S. (2007). Bacterial community structures in MBRs treating municipal wastewater relationship between community stability and reactor performance. Water Research, 41, 627-37. Molina-Muñoz, M.; Poyatos, J. M.; Vílchez, R.; Hontoria, E.; Rodelas, B.; González-López, J. (2007). Effect of the concentration of suspended solids on the enzymatic activities and biodiversity of a submerged membrane bioreactor for aerobic treatment of domestic wastewater. Applied Microbiology Biotechnology, 73, 1441-1451. Naeem, S. and Li, S. B. (1997). Biodiversity enhances ecosystem reliability. Nature 390, 507509. Nuhoglu, A. and Yalcin, B. (2005). Modelling of phenol removal in a batch reactor, Process Biochemistry, 40, 1233-1239. Nubel, U.; Engelen, B.; Felske, A.; Snaidr, J.; Wieshuber, A.; Amann, R. I.; Ludwig, W.; Backhaus, H. (1996). Sequence heterogeneities of genes encoding 16S rRNAs in Paenibacillus polymyxa detected by temperature gradiente gel eletrophoresis. Journal of Bacteriology, 178, 5636-5643. Prieto, M. B.; Hidalgo, A.; Serra, J. L.; Llama, M. J. (2002). Degradation of phenol by Rhodococcus erytropolis UPV-1 immobilized on Biolite in a packed-bed reactor, Journal of Biotechnology, 97, 1-11. Saikaly, P. E. and Oerther, D. B. (2004). Bacterial competition in activated sludge: Theoretical analysis of varying solids retention times on diversity. Microbial Ecology, 48, 274-284. Schneider, R.P. and Tsutiya, M.T. (2001). Membranas Filtrantes para o Tratamento de Água, Esgoto e Água de Reuso. 1 edition, São Paulo, Brasil, Associação Brasileira de Engenharia Sanitária e Ambiental-ABES, 234. Schwartz, M. W.; Brigham, C. A.; Hoeksema, J. D.; Lyons, J. D.; Mills, M. H.; van Mantgeni, P. J. (2000). Linking biodiversity to ecosystem function: implications for conservation ecology. Oecologia, 122, 297-305. Stamper, D. M.; Walch, M.; Jacobs, R. N. (2003). Bacterial Population changes in a membrance bioreactor for graywater treatment monitored by denaturing gradient gel electrophoretic analysis of 16S rRNA gene fragments. Applied Environmental Microbiology, 69, 852-860. Standard Methods for the Examination of Water and Wastewater. (1998). American Public Health Association and American Water Works Association and Water Pollution Control Federation, 20th edition,Washington, DC. Torsvik, V.; Goksoyr, J.; Daae, F. L. (1990). High diversity in DNA of soil bacteria, Applied Environmental Microbiology, 56, 782-787. Ueda, T. and Horan, N. J. (2000). Fate of indigenous bacteriophage in a membrane bioreactor. Water Research, 34, 2151-2159. van der Gaast, C. J.; Reid, J. B. E.; Bailey, M. R. T.; Judd, S. J. (2006). Bacterial diversity is determined by volume in membrane bioreactors. Environment Microbiology, 8, 10481055. Vázquez, I.; Rodríguez, J.; Marañón, E.; Castrillón, L.; Fernández, Y. (2006). Simultaneous removal of phenol, ammonium and thiocyanate from coke wastewater by aerobic biodegradation. Journal of Hazardous Materials, 137, 1773-1780. Viero, A. F.; Melo, T. M.; Torres, A. P. R.; Ferreira, N. M.; Sant’Anna Jr. G. L.; Borges, C. P.; Santiago, V. M. J. (2008). The effects of long-term feeding of high organic loading in

Water Production and Wastewater Treatment, Nova Science Publishers, Incorporated, 2010. ProQuest Ebook Central,

DGGE and 16S rDNA Sequencing Analysis of Bacterial Communities…

43

Copyright © 2010. Nova Science Publishers, Incorporated. All rights reserved.

a submerged membrane bioreactor treating oil refinery wastewater. Journal of Membrane Science, 319, 223-230. Wang, G.C.-Y. and Wang, Y. (1997). Frequency of formation of chimeric molecules as a consequence of PCR coamplification of 16S rRNA genes from mixed bacterial communities. Applied Environmental Microbiology, 63, 4645-4650. Wagner, M. and Loy, A. (2002). Bacterial community composition and function in sewage treatment systems. Current Opinion in Biotechnology, 13, 218-227. Witzig, R.; Manz, W.; Rosenberger, S.; Krugerb, U.; Kraumeb, M.; Szewzyk, U. (2002). Microbiological aspects of a bioreactor with submerged membranes for aerobic treatment of municipal wastewater. Water Research, 36, 394-402. Uygur, A. and Kargi, F. (2004). Phenol inhibition of biological nutrient removal in a four-step sequencing batch reactor. Process Biochemistry, 39, 2123-2128.

Water Production and Wastewater Treatment, Nova Science Publishers, Incorporated, 2010. ProQuest Ebook Central,

Copyright © 2010. Nova Science Publishers, Incorporated. All rights reserved. Water Production and Wastewater Treatment, Nova Science Publishers, Incorporated, 2010. ProQuest Ebook Central,

In: Water Production and Wastewater Treatment Editors: B. Antizar-Ladislao and R. Sheikholeslami

ISBN 978-1-61728-503-5 © 2011 Nova Science Publishers, Inc.

Chapter 4

BIOSORPTION OF CD (II) AND NI (II) FROM AQUEOUS SOLUTIONS BY CYSTOSEIRA INDICA M.M. Montazer-Rahmati a∗, P. Rabbani a and A. Abdolali a a School of Chemical Engineering, College of Engineering, University of Tehran, Tehran, Iran, 11155-4563

Copyright © 2010. Nova Science Publishers, Incorporated. All rights reserved.

ABSTRACT Toxic heavy metal contamination of industrial wastewater is an important environmental problem. Many industries, such as automotive, electroplating, battery manufacturing, mining, electric cable manufacturing, tannery, steel and textile industries release various concentrations of heavy metals such as nickel, cadmium, lead, mercury and copper, etc. into natural waters. These heavy metals are toxic to the aquatic ecosystem and human health even at low concentrations as a result of accumulation in organisms beyond tolerance levels. In this chapter, different methods for uptake of heavy metal ions from aqueous solutions are introduced and biosorption as an effective and low cost method of removing nickel and cadmium ions from synthetic wastewater is discussed. For metal biosorption, brown algae possessing a high content of ioniziable functional groups such as carboxyl, alcohol and amino groups specifically have been previously identified as very promising biosorbents for removing heavy metal ions from synthetic and natural aqueous solutions. The biosorption of brown marine algae, Cystoseira indica, both intact and pretreated by crosslinking with formaldehyde (FA) has been studied. Experimental results have been obtained from batch equilibrium tests. The optimum sorption conditions have been determined and results show that the highest biosorption capacities (19.42 mg/g and 10.06 mg/g for Cd (II) and Ni (II), respectively) onto FA-treated Cystoseira indica (2g/L) are obtained at an optimum pH of 5.5 and 6.0 for Cd (II) and Ni (II), respectively. In the experiments carried out, the initial metal concentration was 0.5 mmol/L, while contact time was about 120 min. One-way ANOVA and one sample t-tests were performed on experimental data to evaluate the statistical significance of biosorption capacities after five sorption and desorption cycles. To describe the biosorption isotherms, the Langmuir, Freundlich, Toth, and Radke-Prausnitz isotherm models were applied. The results fit well to the Radke-Prausnitz and the Toth ∗Corresponding author. Tel.: +98-21-61113217 Fax: +98-21-66957784. E-mail address: [email protected]. Water Production and Wastewater Treatment, Nova Science Publishers, Incorporated, 2010. ProQuest Ebook Central,

46

M.M. Montazer-Rahmati, P. Rabbani and A. Abdolali isotherms for Cd (II) and Ni (II), respectively. The kinetic data were fitted by models including pseudo-first-order and pseudo-second-order. The pseudo-second-order kinetic model describes the biosorption of cadmium and nickel ions quite well.

Copyright © 2010. Nova Science Publishers, Incorporated. All rights reserved.

1. INTRODUCTION Heavy metals are major pollutants that are released into the natural environment as wastes from industrial processes and have resulted in a number of environmental problems. Cadmium and nickel are among the heavy metals that are used widely. Cadmium is extremely toxic even at low concentrations [1-3]. The industrial uses of these metals are in plastics, alloy preparation, wood treating, and covered metalworking operations, paint pigments, mining, ceramics, electroplating and batteries; therefore, they pose a significant potential hazard to the environment and human beings [4]. Since heavy metals cannot be annihilated in the natural environment, there is a need for new technologies that can remove and recover heavy metals from wastewaters [3, 5]. At present, a number of technologies such as adsorption, chemical precipitation, electroplating, and ion exchange are used for the treatment of wastewater streams containing heavy metals [2, 4]. But when heavy metal concentrations in the wastewater are low, these methods are not effective and are very expensive when the concentration of heavy metals in the effluent is very low [6]. Biosorption has been found to be a feasible economical alternative as a wastewater treatment process for metal removal [68]. It has been reported that the biomass of brown algae has a metal binding capacity higher than the other types of biosorbents such as fungi, bacteria and yeast [1, 9]. The mechanism of binding metal ions by algal biomass may depend on the species and ionic charges of the metal ions, the algal organisms, the chemical composition of the metal ion solution and other external environmental factors such as pH and temperature. The purpose of this chapter is to study cadmium (II) and nickel (II) removal from aqueous solutions by both intact and pre-treated brown algae from the Persian Gulf. The biosorption process was analyzed through batch experiments. The influence of pH, modification and contact time was studied. Also sorption and desorption experiments were carried out. In order to investigate the sorption behaviour, various kinetic and isotherm models were studied.

1.1. Physico-Chemical Methods for Heavy Metal Removal Different treatment techniques for wastewater containing heavy metals have been developed in recent years both to decrease the amount of wastewater produced and to improve the quality of the treated effluent. Although various treatments such as chemical precipitation, coagulation–flocculation, floatation, ion exchange and membrane filtration can be employed to remove heavy metals from contaminated wastewater, they have their inherent advantages and limitations in application [10].

1.1.1. Chemical Precipitation As metals enter the treatment process, they are in a stable, dissolved aqueous form and are unable to form solids. The goal of metal treatment by hydroxide precipitation is then to adjust the pH (hydroxide ion concentration) of the water so that the metals will form insoluble

Water Production and Wastewater Treatment, Nova Science Publishers, Incorporated, 2010. ProQuest Ebook Central,

Copyright © 2010. Nova Science Publishers, Incorporated. All rights reserved.

Biosorption of Cd (II) and Ni (II) from Aqueous Solutions...

47

precipitates. Once the metals precipitate and form solids, they can then be removed easily, and the water, now having a low metal concentration, can be discharged. Metal precipitation is primarily dependent upon two factors: the concentration of the metal, and the pH of the water. Heavy metals are usually present in wastewaters in dilute quantities (1-100 mg/L) and at neutral or acidic pH values (< 7.0). Both of these factors are disadvantageous with regard to metal removal. However, when one adds caustic to water containing dissolved metals, the metals react with hydroxide ions to form metal hydroxide solids. Some variations in the exact values of the metal concentrations will occur due to the presence of other substances in the wastewater. Compounds such as cyanide or ammonia can inhibit precipitation of metals, and limit their removal to the point where discharge limits can be exceeded. Also, not all metals have the same minimum solubility. Therefore, in a wastewater containing various metals, as a general rule, pH should be adjusted to an average value of approximately 9. However, the metals now exist in another phase or state (i.e., as small solid particles). Metal removal is not complete until these metal solids are physically removed from the wastewater, typically by subsequent sedimentation and filtration. Metal removal occurs through the use of several unit operations: In the first step the wastewater enters the rapid mix unit. The goal of the rapid mix operation is to first raise the wastewater pH to form metal hydroxide particles, as discussed above. After the addition of caustic, the next step is to add aluminium or iron salts, or organic polymers (coagulants) directly into the wastewater. These polymers attach to the solid metal particles. The small metal hydroxide particles become entangled in these polymers, causing the particle size to increase (form flocs), thus promoting the settling process. Once particles become enmeshed in the polymer, they are allowed to settle so that they can be removed from the wastewater. The particles settle since they are heavier than water. Settling occurs in the sedimentation tanks. Sedimentation tanks, in contrast to rapid mixing units, are designed to have no mixing and to provide a calm flow for settling. Water emerging from the sedimentation basin is routed to the filtration unit. The filtration unit is designed to trap those particles that did not settle in the sedimentation basin (because they were too small) or did not have sufficient time to settle and were carried out of the basin. Water entering the filtration unit is passed through silica sand, diatomaceous earth, carbon, or cloth to capture the remaining metal hydroxide particles. Metal particles stick to the filtering material and are removed from the water. Filtration completes the metal treatment process. Only now should the pH be reduced for discharge, if necessary, or the pH can now be adjusted for water reuse. The solids produced in the sedimentation stage (and possibly solids from filtration) are denoted as a sludge and periodically removed. In diatomaceous earth and fiber filters, the entire filter media (diatomaceous earth, filter cartridge) is dumped with the captured metal hydroxide solids. This sludge may be sent to a dewatering stage to remove excess water and leave only solids. The water from the dewatering stage may not be completely free of metals and should be piped to the rapid mix tank.

Water Production and Wastewater Treatment, Nova Science Publishers, Incorporated, 2010. ProQuest Ebook Central,

48

M.M. Montazer-Rahmati, P. Rabbani and A. Abdolali

Copyright © 2010. Nova Science Publishers, Incorporated. All rights reserved.

The sludge now contains the precipitated metal hydroxide solids, made up of identifiable quantities of heavy metals, which are regulated according to applicable guidelines. The solids produced from heavy metal wastewater treatment must then be disposed of as a hazardous waste. The applicability of hydroxide precipitation in a closed system was studied to remove Cd (II) and Cu (II) ions from synthetic wastewater [7]. Inorganic cations (Ca (II) and Na) were employed as ligand-sharing agents for EDTA (Ethylene Diamine Tetraacetic Acid) and NTA (Nitrilo Triacetic Acid). It was reported that Ca (II) was the only cation that could effectively bond with both ligands to form hydroxide precipitates of the complexed metals. Another study by Papadopoulos et al. [11] pertains to experiments carried out for Ni (II) removal from a low-strength real wastewater using hydroxide precipitation.

1.1.2. Membrane Filtration Membrane filtration has received considerable attention for the treatment of inorganic effluents, since it is capable of removing not only suspended solid and organic compounds, but also inorganic contaminants such as heavy metals. Depending on the size of the particles that can be retained, various types of membrane filtration such as ultra-filtration (UF), nanofiltration (NF) and reverse osmosis can be employed for heavy metal removal. UF utilizes permeable membranes to separate heavy metals, macromolecules and suspended solids from inorganic solutions based on the pore size (5–20 nm) and molecular weight of the compounds to be separated (1000–100,000). These unique capabilities enable UF to allow the passage of water and low-molecular weight solutes, while retaining the macromolecules, having a size larger than the pore size of the membrane [12]. Depending on the membrane characteristics, UF has a removal efficiency of more than 90% with a metal concentration ranging from 10 to 112 mg/L at pH values ranging from 5 to 9.5 and at pressures in the range of 2–5 bars. Although UF needs a lower driving force and smaller space, this method has high operating costs. 1.1.2.1. Reverse Osmosis Reverse osmosis (RO) is a separation process that uses pressure to force a solution through a membrane that retains the solute on one side and allows the pure solvent to pass to the other side. More formally, it is the process of forcing a solvent from a region of high solute concentration through a membrane to a region of low solute concentration by applying a pressure in excess of the osmotic pressure. This is the reverse of the normal osmosis process, which is the natural movement of solvent from an area of low solute concentration, through a membrane, to an area of high solute concentration when no external pressure is applied. The membrane is semi-permeable, meaning it allows the passage of solvent but not of solute. The membranes used for reverse osmosis have a dense barrier layer in the polymer matrix where most separation occurs. In most cases the membrane is designed to allow only water to pass through this dense layer while preventing the passage of solutes (such as salt ions). This process requires that a high pressure be exerted on the high concentration side of the membrane [13, 14].

Water Production and Wastewater Treatment, Nova Science Publishers, Incorporated, 2010. ProQuest Ebook Central,

Biosorption of Cd (II) and Ni (II) from Aqueous Solutions...

49

Copyright © 2010. Nova Science Publishers, Incorporated. All rights reserved.

1.1.2.2. Nanofiltration Nanofiltration has unique properties compared with UF and RO membranes. Its separation mechanism involves steric (sieving) and electrical (Donnan) effects. A Donnan potential is created between the charged anions in the NF membrane and the co-ions in the effluent to reject the latter [15]. The significance of this membrane lies in its small pore and membrane surface charge, allowing charged solutes smaller than the membrane pores to be rejected along with the bigger neutral solutes and salts. Ahn et al. [16] worked on Ni (II) ion removal from real electroplating wastewater using NTR-7250 membranes in order to evaluate polyvinyl alcohol as the skin materials of the NF membrane. They found that the removal of Ni (II) depended on the initial metal concentrations and the applied pressure. Moreover, from their experiments, it was observed that beyond a pressure of 2.9 bars, the removal of Ni (II) did not improve with increasing pressure, therefore it was suggested that 2.9 bars was the optimum pressure for NF application to remove Ni (II) ions from wastewater. A comparative study of Cu (II) and Cd (II) removal from synthetic wastewater using nanofiltration (NF) and reverse osmosis (RO) has been conducted [17]. At the same initial metal concentration, while NF was capable of removing only 90% of Cu (II) and 97% of Cd (II); 98% of Cu (II) removal and 99% of Cd (II) removal could be achieved using RO. These results indicate that both types of membrane filtration are effective for metal removal from contaminated wastewater. Nevertheless, NF requires lower pressures than RO, making NF more preferable due to its lower operating costs. Depending on the membrane specifications, NF can effectively remove heavy metal ions with a concentration of approximately 2000 ppm at a wide pH range of 3–8 and at pressures of 3–4 bars. However, NF is less intensively investigated than UF and RO for the removal of heavy metals [18]. 1.1.3. Coagulation–Flocculation Coagulation–flocculation can be employed to treat wastewater. Principally, the coagulation process destabilizes colloidal particles by adding a coagulant and results in sedimentation. To increase the particle size, coagulation is followed by the flocculation of the unstable particles into bulky floccules [19]. The general approach for this technique includes pH adjustment and involves the addition of ferric/alum salts as the coagulant to overcome the repulsive forces between particles [20]. In general, coagulation–flocculation can treat inorganic effluents with a metal concentration of less than 100 mg/L or higher than 1000 mg/L. Improved sludge settling, dewatering characteristics, bacterial inactivation capability, sludge stability are reported to be the major advantages of lime-based coagulation. In spite of its many advantages, the coagulation–flocculation process has disadvantages such as high operating costs due to chemical consumption. The increased volume of sludge generated from the coagulation–flocculation process may hinder its adoption as a global strategy for wastewater treatment. This can be attributed to the fact that the toxic sludge must be converted into a stabilized product to prevent heavy metals from leaking into the environment [21]. To overcome such problems, electro-coagulation may be a better alternative than the conventional coagulation, as it can remove the smallest colloidal particles and produce just a small amount of sludge [22,23]. However, this technique also creates a floc of metallic

Water Production and Wastewater Treatment, Nova Science Publishers, Incorporated, 2010. ProQuest Ebook Central,

50

M.M. Montazer-Rahmati, P. Rabbani and A. Abdolali

Copyright © 2010. Nova Science Publishers, Incorporated. All rights reserved.

hydroxides, requiring further purification and making the recovery of valuable heavy metals impossible.

1.1.4. Ion-Exchange Ion exchange is a reversible chemical reaction wherein an ion (an atom or molecule that has lost or gained an electron and thus acquired an electrical charge) from solution is exchanged for a similarly charged ion attached to an immobile solid particle. These solid ion exchange particles are either naturally occurring inorganic zeolites or synthetically produced organic resins. The synthetic organic resins are the predominant type used today because their characteristics can be tailored to specific applications. An organic ion exchange resin is composed of high-molecular-weight polyelectrolytes that can exchange their mobile ions for ions of similar charge from the surrounding medium. Each resin has a distinct number of mobile ion sites that set the maximum quantity of exchanges per unit of resin. Ion exchange reactions are stoichiometric and reversible, and in that sense they are similar to other solution phase reactions. Unlike chemical precipitation, ion exchange does not present any sludge disposal problems [24], thus lowering the operating costs for the disposal of the residual metal sludge. Other advantages of ion exchange include convenience for fieldwork since the required equipment is portable, the speciation results are reliable and the experiments can be done quickly. Resins also have certain ligands that can selectively bond with certain metal cations, making ion exchange easy to use and less time-consuming [25]. Despite these advantages, ion exchange also has some limitations in treating wastewater laden with heavy metals. Prior to ion exchange, appropriate pretreatment systems for secondary effluent such as the removal of suspended solids from wastewater are required. In addition, suitable ion exchange resins are not available for all heavy metals, while the capital and operating costs are high [26]. 1.1.5. Adsorption Recently, adsorption has become one of the alternative treatment techniques for wastewater laden with heavy metals [27]. Basically, adsorption is a mass transfer process by which a substance is transferred from the liquid phase to the surface of a solid, and becomes bound by physical and/or chemical interactions [10]. Due to its large surface area, high adsorption capacity and surface reactivity, adsorption using activated carbon can remove metals from inorganic effluents. Activated carbons are more effective in the removal of heavy metals due to some specific characteristics that enhance the use of activated carbon for the removal of contaminants including heavy metals from water supplies and wastewater [28]. Natural materials that are available in large quantities or certain waste products from industrial operations [29] and agricultural by-products [30, 31] may have potential as inexpensive adsorbents. Generally, adsorbents can be assumed as low cost if they require little processing, are abundant in nature, or are a by-product or waste material from another industry [32]. In general, technical applicability and cost-effectiveness are the key factors that play major roles in the selection of the most suitable adsorbent to treat inorganic effluents.

Water Production and Wastewater Treatment, Nova Science Publishers, Incorporated, 2010. ProQuest Ebook Central,

Biosorption of Cd (II) and Ni (II) from Aqueous Solutions...

51

Copyright © 2010. Nova Science Publishers, Incorporated. All rights reserved.

1.1.6. Evaporation Evaporation is a process in which the effluent is heated as much as possible and the concentration of solids in it increases as a result of partial evaporation. The steam produced can be used for power generation or changed to liquid to be reused again. For evaporation of effluents, a large place is needed leading to high investment costs. Besides the high fixed and operating costs, this process is very time consuming. In this method, metals are not separated individually, so a concentrated sludge consisting of all the metals is produced. Therefore, an auxiliary method such as ion-exchange is required for separating particular metals. However, all the methods discussed so far have several disadvantages; which include incomplete metal removal and toxic sludge generation or other waste products that require disposal. On the other hand, the use of artificially prepared ion exchange resins is effective, but too expensive to be applied on industrial scale. These techniques, apart from being expensive, have other disadvantages such as high reagent and energy requirements. For this reason, the potential of a new method of removing heavy metals using biosorbents may provide an important breakthrough [33]. In Table 1, the different heavy metal uptake methods are compared and their advantages and disadvantages are presented. 1.1.7. Biosorption Heavy metal removal from wastewater by the adsorption process has a short history compared to other methods of water purification [34]. The potential of algae biosorption is beginning to be recognized both for the recovery of valuable metals and also for reducing pollution [35]. The potential of adsorption was first observed by Lowitz in 1785 and was soon applied as a process for removal of color from sugar during refining. In the second half of the nineteenth century, American water treatment plants used nonactivated charcoal filters for water purification. In 1929 the first granular activated carbon (GAC) units for treatment of water supplies were constructed in Hamm, Germany, and in 1930 at Bay City, Michigan. Inasmuch as the cost of activated carbon was so high as to limit its use in adsorption, a number of studies for a low-cost and readily available adsorbent led to the investigation of agricultural and biological materials as potential metal sorbents. Biosorption is the ability of certain types of microbial biomass to adsorb heavy metals from aqueous solutions. Biosorbents may be viewed as natural ion-exchange materials that primarily contain weakly acidic and basic groups [34]. In recent decades, several investigators [9, 36-38] have reported the ability of metal removal of natural materials such as rice, coconut husks, peat moss, peanut skin, waste tea leaves, sugar cane bagasse, carrot and orange peels, and so on. Many studies have shown that nonliving plant biomass materials are effective for the removal of trace metals from the environment [39]. Live or dead biomass can be used as an adsorbent for removal of toxic metal ions from aqueous solutions. The efficiency of dead cells in biosorbing metal ions may be greater, equivalent to, or less than that of living cells and may depend on factors such as biosorbent properties, the pre-treatment method used, and the type of metal ion being studied.

Water Production and Wastewater Treatment, Nova Science Publishers, Incorporated, 2010. ProQuest Ebook Central,

M.M. Montazer-Rahmati, P. Rabbani and A. Abdolali

52

Copyright © 2010. Nova Science Publishers, Incorporated. All rights reserved.

Table 1. Summary of the applicability of various physico–chemical techniques for wastewater treatment [10] Type of treatment

Target of removal

Advantages

Disadvantages

Chemical precipitation

Heavy metals, divalent metals

Low capital cost, simple operation

Coagulation– flocculation

Heavy metals and suspended solids

Evaporation

Heavy metals and suspended solids

Shorter time to settle out suspended solids, improved sludge settling Using the generated steam in utilities

Sludge generation, extra operational cost for sludge disposal Sludge production, extra operational cost for sludge disposal

Ion exchange

Dissolved compounds, cations/anions

No sludge generation, less time consuming

Sludge production, Subsequent treatments are required to improve the removal efficiency of heavy metals, high cost Not all ion exchange resins are suitable for metal removal, high capital cost

Ultrafiltration

High molecular weight compounds (1000–10000)

Smaller space requirement

High operational cost, prone to membrane fouling

Nanofiltration

Sulphate salts and hardness ions such as Ca(II) and Mg(II)

Lower pressure than RO (7–30 bar)

Costly, prone to membrane fouling

Reverse osmosis

Organic and inorganic compounds

High rejection rate, able to withstand high temperature

High energy consumption due to high pressure required (20– 100 bar), susceptible to membrane fouling

Biosorption

Heavy metals

Efficient in very low concentration of heavy metals in dead and alive biomasses, low cost, simple operation and less time consuming

Subsequent treatments are required to improve the removal efficiency of heavy metal for tapped water

The use of dead cells over live cells offers the following advantages: • • • •

The metal removal system is not subject to toxicity limitations There is no requirement for growth media and nutrients The biosorbed metal ions can be easily desorbed and biomass can be reused and The dead biomass-based treatment systems can be subjected to the traditional adsorption models in use.

Water Production and Wastewater Treatment, Nova Science Publishers, Incorporated, 2010. ProQuest Ebook Central,

Copyright © 2010. Nova Science Publishers, Incorporated. All rights reserved.

Biosorption of Cd (II) and Ni (II) from Aqueous Solutions...

53

As a result, use of dead biomass has been a preferred choice of numerous studies on biosorption of toxic metal ions from aqueous solutions [40]. In biosorption, the metal binding on the active surface sites takes place via various mechanisms such as complexation, coordination, chelating, physical and/or chemical adsorption, ion-exchanging and microprecipitation. A large number of microorganisms like bacteria, fungi, yeast and marine algae constitute potential biosorbents. A number of studies [41-45] demonstrated that among the many possible biosorbents, marine algae especially brown algae are excellent biosorbents for metals. Among biosorbent materials, algae have proved to be both economical and eco-friendly, as they are abundantly available, have regeneration and metal recovery potential, lower volume of chemical and/or biological sludge to be disposed of, high efficiency in dilute effluents and high surface area to volume ratio. Their use provides a cost effective solution for industrial wastewater management. The brown algae are an important assembly of plants that are classified in about 265 genera with more than 1500 species [46]. The biorecovery of heavy metals can be affected by physico-chemical parameters of the solution such as pH, ion strength and temperature and by other characteristics of the biomass like concentration, presence of organic and inorganic functional groups and chemical modification (pre-treatment) [37,47-50]. Even though the brown algae are one of the most commonly used biosorbents for heavy metal recovery, they may cause a secondary pollution due to organic substances release, for example, alginate dissolving from biosorbents during the biosorption process. This phenomenon may hinder their industrial application. Also the leaching of some adsorbtive components may lead to a loss of biosorption capacity [51]. Some studies [51-53] have shown that surface modification by calcium chloride, formaldehyde and glutaraldehyde can prevent leaching of adsorptive components from biomass and increase the stability of the biosorbent material. Polyethylene imine (PEI) is well known for its metal chelating characteristics due to the presence of a large number of amine groups in a molecule and it is often used to modify the adsorbent surface to increase the adsorption capacity [9]. The results obtained indicate that the pretreated algae have a better potential for the removal of heavy metals from wastewaters [54]. The carboxylic groups are generally the most abundant acidic functional groups in brown algae. They constitute the highest percentage of titrable sites (typically greater than 70%) in dried brown algal biomass. The adsorption capacity of the algae is directly related to the presence of these sites on the alginate polymer, which itself comprises a significant component (up to 40% of the dry weight) of the dried seaweed biomass. Furthermore, the majority of metals of interest (i.e. Cd2+, Co2+, Cu2+, Zn2+, Ni2+ and Pb2+) display maximal or near maximal sequestration at pHs near the apparent dissociation constant of carboxylic acids observed in brown algal biomass (near 5). The role of carboxylic groups in the adsorption process has been clearly demonstrated by a reduction in heavy metal ion removal by dried blown algae. In general, biosorption by brown algae is efficient at very low concentrations of heavy metals using dead and living biomass, is low cost, is simple and less time consuming in operation.

Water Production and Wastewater Treatment, Nova Science Publishers, Incorporated, 2010. ProQuest Ebook Central,

M.M. Montazer-Rahmati, P. Rabbani and A. Abdolali

54

2. DISCUSSSION 2.1. Materials and Preparation of Pre-Treated Biomass Stock metal solutions were prepared by dissolving cadmium and nickel nitrate, Cd(NO3)2.4H2O and Ni(NO3)2.6H2O (Merck, Germany) in double distilled water (DDW). All the reagents were obtained from Merck (Germany). Samples of brown marine algae, Cystoseira indica, were collected from the coastal areas of the Persian Gulf, Iran. They were washed with tap water to remove sand and other impurities and were sun-dried for 48 h. The biomass was crushed and sieved (RETSCH AS200, Germany) to a particle size of 0.5-1.0 mm and then washed with DDW and dried in an oven (Heraeus CH 20P, Germany) at 80˚C for 24 h. This material will be referred to as intact biomass. Surface modification by FA can prevent leaching of the adsorptive components from biomass and increase the stability of the biosorbent material [51]. For preparation of the pre-treated algae by 10% FA, 10 g/L of unwashed biomass was stirred with the solution at 25˚C and at 100 rpm for 1 h. The pre-treated biomass was washed several times with deionized water and then dried in the oven at 60˚C overnight. The dried biomass was then sieved to get a uniform particle size of 0.5-1.0 mm.

Copyright © 2010. Nova Science Publishers, Incorporated. All rights reserved.

2.2. Batch Biosorption Experiments The biosorption experiments were carried out in a series of 100ml Erlenmeyer flasks containing metal nitrate solutions with a metal concentration of 0.5 mmol/L of each metal ion, 0.06g (2g/L) of biomass and a buffer solution to adjust the pH. Contents of the Erlenmeyer flasks were shaken in the incubator shaker (INFORS multitron, Switzerland) at 25˚C and at 150 rpm for 2 h. After this time, the biomass was separated from the solution using a centrifuge (Sigma 203, Germany) at 4000 rpm for 15 min. The concentration of metal in the remaining solution was measured by an ICP-OES (Inductively Coupled Plasma-Optical Emission Spectrometer) (Perkin Elmer-Optima 2100, USA). The amount of cadmium and nickel adsorbed q (mg/g), by FA-treated Cystoseira indica was calculated using the following equation [45]: q=

v (C i − C f

)

m

(1)

where Ci and Cf (mg/L) are the initial and final metal ion concentrations in the solution, respectively, v (L) is the solution volume and m (g) is the mass of the biosorbent.

2.3. Effect of pH Zeta potential measurements showed that the immobilized biosorbent was negatively charged in the pH range of 3.0-8.0 [55]. An additional possible explanation why sorption increases with increasing pH is that the solubility of many metals in solution decreases with

Water Production and Wastewater Treatment, Nova Science Publishers, Incorporated, 2010. ProQuest Ebook Central,

Biosorption of Cd (II) and Ni (II) from Aqueous Solutions...

55

Copyright © 2010. Nova Science Publishers, Incorporated. All rights reserved.

increasing pH. A further possible explanation of increasing sorption with increasing pH is that hydrolyzed species have a lower degree of hydration, i.e., less energy is necessary for removal or reorientation of the hydrated water molecules upon binding. With a further increase of pH (6–9), the solubility of metals decreases enough for precipitation to occur. This should be avoided during sorption experiments because otherwise distinguishing between sorption and precipitation metal removal becomes difficult [56]. Experiments were conducted with FA-treated Cystoseira indica for cadmium (II) and nickel (II) biosorption in batch systems. The adsorption characteristics of cadmium (II) and nickel (II) at various pH values in the range of 2.5-7.0 were examined. Experiments were not conducted beyond a pH of 7.0 to avoid metal precipitation as mentioned before. As shown in Fig. 1, biosorption of these metals are highly pH-dependent and adsorption of cadmium and nickel is highest at pH values of 5.5 and 6.0, respectively, and then decreases as the pH increases. Therefore, it is obvious that modification causes an increase in metal biosorption. The pH dependency of metal sorption is explained considering the nature of biosorbents. The cell wall of brown algae mainly contains a large number of surface functional groups, among which carboxyl is generally the most abundant acidic functional group. At low pH values, cell wall ligands are protonated and compete significantly with metal ions for binding. With increasing pH, more ligands such as amino and carboxyl groups would be exposed leading to attraction between these negative charges and metal ions. In addition, at low pH values, the concentration of H3O+ far exceeds that of metal ions and hence occupies the binding sites on the cell walls, leaving metal ions unbound. When the pH values are increased, the competing effect of H3O+ decreases and positively charged metal ions take up the free binding sites, therefore metal sorption increases [52, 54].

Figure 1. Effect of pH on Cd (II) and Ni (II) biosorption by intact and FA-treated C. indica biomass (Cd and Ni conc.: 0.5 mmol/L; contact time: 2 h; T: 25˚C; biomass dosage: 2g/L).

2.4. Sorption and Desorption Experiments Because of economic reasons, regeneration of biomass seems to be necessary. For desorption of Cd (II) and Ni (II) ions from biomass, three desorbing agents 0.1M CaCl2, 0.1M

Water Production and Wastewater Treatment, Nova Science Publishers, Incorporated, 2010. ProQuest Ebook Central,

56

M.M. Montazer-Rahmati, P. Rabbani and A. Abdolali

Copyright © 2010. Nova Science Publishers, Incorporated. All rights reserved.

CH3COOH and 0.1M NaCl have been used [35]. The biomass biosorption capacity was tested in five repeated cycles at an optimum pH of 5.5 and 6.0 for Cd (II) and Ni (II), respectively. Figure 2 indicates that the biosorption capacity of the biomass is good after 5 cycles. To test the significance and adequacy of the model, statistical testing of the model in the form of analysis of variance (ANOVA) and the one-sample t-test were done. For a 5% level of significance, F and critical F values are 0.55 and 3.89 for Cd (II) (p value = 0.59) and 2.32 and 3.80 for Ni (II) (p value = 0.141), respectively. These values show the variation of sorption capacities among the three desorbing agents not to be significant after 5 cycles. For a 5% level of significance, t values for NaCl, CH3COOH and CaCl2 desorbing agents are 0.61, 1.66 and 4.44 for Cd (II) and 0.85, 1.42 and 3.91 for Ni (II), respectively and the critical t is 3.18 for both. Results show that the effect of CaCl2 on the biosorbent is significant and CaCl2 causes an increase in the sorption capacity. Hence, CaCl2 is recommended as a good desorbent for elution and desorption of cadmium and nickel from the biosorbent. Of course the much lower cost of NaCl should also be taken into consideration.

Figure 2. Effect of desorping agents on metal biosorption onto FA-treated C. indica biomass at optimum pH (T: 25˚C; sorption contact time: 45 min; desorption contact time: 20 min; 5 cycles) (a) Cd (II) conc.: 0.05 mmol/L and (b) Ni (II) conc.: 1 mmol/L.

2.5. Effect of Initial Concentrations The adsorption concentrations increased with an increase in the initial metal ion concentration. This was most probably due to an increase in the initial ion concentrations providing more chance for biosorbent contact.

2.6. Effect of Biosorbent Dosage Biosorbent dosage seemed to have a great influence on the biosorption process. The dosage of biomass added into the solution determines the number of binding sites available for adsorption.

Water Production and Wastewater Treatment, Nova Science Publishers, Incorporated, 2010. ProQuest Ebook Central,

Biosorption of Cd (II) and Ni (II) from Aqueous Solutions...

57

Copyright © 2010. Nova Science Publishers, Incorporated. All rights reserved.

Figure 3. Effect of biomass dosage on biosorption of Ni (II) onto FA-treated C. indica biomass (Ni conc.: 1.0 mmol/L; Temperature: 25˚C; Contact time: 3h; pH: 6.0).

Figure 4. Effect of biomass dosage on biosorption of Cd (II) onto FA-treated C. indica biomass (Cd conc.: 0.5 mmol/L; Temperature: 25˚C; Contact time: 3h; pH: 5.5).

Linear regression has been frequently used to evaluate the model parameters; however, the non-linear method is to be preferred because this method is more suited to the error structure of the data. The error of the estimates of the parameters can only be calculated consistently from a non-linear fitting procedure [57]. The effect of biomass dosage on Ni and Cd ion removal is indicated in Figs. 3 and 4.

2.7. Biosorption Isotherms For optimization of the biosorption process design, it is necessary to acquire the appropriate correlation for the equilibrium curve. In this chapter, the relationship between metal biosorption capacity and metal concentration at equilibrium has been described by two-

Water Production and Wastewater Treatment, Nova Science Publishers, Incorporated, 2010. ProQuest Ebook Central,

M.M. Montazer-Rahmati, P. Rabbani and A. Abdolali

58

parameter sorption isotherm models: Langmuir and Freundlich and three-parameter isotherm models: Toth and Radke-Prausnitz. The results of the other models described herein are not given in these tables. The constants, residual root mean square error (RMSE), chi-square test (χ2) and correlation coefficient (R2) of two and three–parameter models are given in Tables 2 and 3, respectively. All model parameters are evaluated by non-linear regression using the MATLAB® software. Furthermore, RMSE, χ2 and R2 are used to measure the goodness of fit. RMSE and χ2 can be defined as: RMSE = (

m 1 ) ∑ (q i ,exp − q i , cal ) 2 m − 2 i =1

(q i ,exp − q i , cal )2 q i ,exp i =1 m

Copyright © 2010. Nova Science Publishers, Incorporated. All rights reserved.

χ2 = ∑

(2)

(3)

where qi,exp and qi,cal are the observation from the batch experiment and the estimate from the isotherm for the corresponding qi,exp, respectively, and m is the number of observations in the experimental isotherm. The smaller RMSE value indicates a better curve fitting, moreover, if the data from the model are similar to the experimental results, χ2 will be a small number [58]. Two-parameter isotherms: Several mathematical models have been developed to quantitatively express the relationship between the extent of sorption and the residual solute concentration. The most widely used model is the Langmuir adsorption isotherm model. Virtually all theoretical treatments of adsorption phenomena are based on or can be readily related to the analysis developed by Langmuir [8]. This model provides a simple mechanistic picture of the adsorption process and gives rise to a relatively simple mathematical expression, which follows: qe = q m

bC e 1 + bC e

(4)

where qm is the maximum metal biosorption and b (kadsorption/kdesorption) is the adsorption equilibrium constant (L/mg). These constants are related to monolayer adsorption capacity and energy of adsorption, respectively [58]. Maximum monolayer adsorption capacity (qm) was obtained as 19.56 mg/g and 16.17 mg/g for Cd (II) and Ni (II) sorption, respectively. The b values of Cd (II) and Ni (II) biosorption are estimated from the isotherms to be 0.05 L/mg and 0.04 L/mg, respectively. The term qm is presumed to represent a fixed number of surface sites in the sorbent, and it should, therefore, be constant and temperature-independent. This term is only determined by the nature of the sorbent. The Langmuir model was originally derived for the adsorption of gases onto activated carbon with several assumptions:

Water Production and Wastewater Treatment, Nova Science Publishers, Incorporated, 2010. ProQuest Ebook Central,

Biosorption of Cd (II) and Ni (II) from Aqueous Solutions...

59

1. There is a limited area available for adsorption, 2. The adsorbed solute material on the surface is only one molecule in thickness, and 3. Adsorption is reversible and an equilibrium condition is achieved. The two parameters of the Langmuir isotherm reflect the maximum uptake and the affinity of the component for the sorbent. As it was shown for the biosorption of heavy metals, metal ion removal is eventually limited by the fixed number of active sites and a resulting plateau can be observed. This phenomenon is well depicted by Langmuir isotherms. The Freundlich equation is an empirical expression based on biosorption on a heterogeneous surface. The Freundlich model is recognized as the earliest empirical equation and is shown to be consistent with an exponential distribution of active centers, characteristic of heterogeneous surfaces. This isotherm model is presented as follows [58]:

Copyright © 2010. Nova Science Publishers, Incorporated. All rights reserved.

qe = k FC e

1

(5)

n

where kF (L/g) is the Freundlich constant which is an important constant indicating a relative measure for adsorption efficiency and n is the Freundlich exponent. The magnitude of n is an indication of the favorability of adsorption. Values of n larger than 1 indicate favorable adsorption [59]. It is assumed that the stronger binding sites are occupied first. The kF and n values in the Freundlich equation are calculated as 0.007 L/g, 0.56 L/g, and 0.52 and 1.27 for Cd (II) and Ni (II), respectively. From Table 2, it is apparent that equilibrium data of Cd (II) biosorption are fitted well by the Freundlich isotherm, the values of R2 of the Freundlich and Langmuir isotherm models are 0.996 and 0.897, respectively, and for Ni (II) the Langmuir isotherm gave a better fit than the Freundlich isotherm based on the values of R2 of the Langmuir isotherm model (0.997) which is higher than that of the Freundlich isotherm (0.995) although not significantly so (only the third decimal is different). Since the Freundlich and Langmuir isotherm models do not provide any clues about the mechanism of biosorption, the equilibrium data are tested with the Dubinin–Radushkevich isotherm model. Dubinin and Radushkevich have reported that the characteristic biosorption curve is related to the porous structure of the sorbents. The Dubinin–Radushkevich (D-R) equation is generally expressed as follows: q e = q DR exp ( − B DR ε DR 2 )

(6)

⎛

(7)

ε DR = RT ln ⎜1 + ⎝

1 ⎞ ⎟ Ce ⎠

where εDR, the Polanyi potential, is a constant related to the biosorption energy, R is the gas constant (8.314 kJ/mol) and T is the absolute temperature (K). qDR and BDR are the D-R isotherm constants in mg/g and mol2/kJ2, respectively. The Polanyi sorption theory assumes a fixed volume of sorption space close to the sorbent surface and the existence of a sorption potential over these spaces. The mean free energy of biosorption (E) can be calculated from the following equation:

Water Production and Wastewater Treatment, Nova Science Publishers, Incorporated, 2010. ProQuest Ebook Central,

M.M. Montazer-Rahmati, P. Rabbani and A. Abdolali

60 E=

1

(8)

2 BDR

Three-parameter isotherms: The Toth isotherm, derived from the potential theory, is expressed as: qe =

q m,T bT C e

[1 + (b C ) ] 1

T

e

nT

nT

(9)

where bT is the Toth model constant and nT is the Toth model exponent. It is obvious that for nT=1 this isotherm reduces to the Langmuir equation. The Toth model constant bT and exponent nT values for Cd (II) and Ni (II) are found to be 0.10 L/mg, 3.67 L/mg, and -0.16 and -0.21, respectively. The Radke-Prausnitz isotherm can be represented as [60]:

Copyright © 2010. Nova Science Publishers, Incorporated. All rights reserved.

qe =

a R − P rR − p C e

βR−P

a R − P + rR − P C e

β R − P −1

(10)

where aR-P and rR-P are Radke-Prausnitz model constants and βR-P is the Radke-Prausnitz model exponent. The Radke-Prausnitz isotherm constants, aR-P and rR-P for Cd (II) and Ni (II) are calculated as -0.27 mg/g and 1.02 mg/g, and 0.0004 L/g and 0.37 L/g, respectively. As the results shown in Table 3 indicate, the experimental results of Cd (II) biosorption are fitted quite well by the Radke-Prausnitz isotherm model and Ni (II) biosorption data are correlated well by the Toth model as confirmed by small values of RMSE and χ2, and R2 values close to 1.0. Khan's isotherm is as follows: qe =

qm, K bK Ce

(1 + bK Ce )

aK

(11)

where bK is the Khan model constant and aK is the Khan model exponent. The Redlich–Peterson isotherm approximates the Henry’s law at low sorbate concentrations, and at high concentrations it behaves like the Freundlich isotherm. It is given as [58]: qe =

K RP C e 1 + aRP C e β RP

(12)

where KRP and aRP are the Redlich–Peterson model constants in L/g and L/mg, respectively and βRP is the Redlich–Peterson model exponent lying between 0 and 1.

Water Production and Wastewater Treatment, Nova Science Publishers, Incorporated, 2010. ProQuest Ebook Central,

Biosorption of Cd (II) and Ni (II) from Aqueous Solutions...

61

Table 2. Isotherm constants of the two-parameter models for Cd (II) and Ni (II) biosorption on FA-treated C. indica biomass at 25˚C Metal

Equilibrium model

Cadmium

Nickel

Langmuir qm (mg/g) b (L/mg) R2 RMSE χ2

19.56 0.05 0.897 7.75 18.74

16.17 0.04 0.997 2.38 2.57

Freundlich KF n R2 RMSE χ2

0.007 0.52 0.996 5.62 10.77

0.56 1.27 0.995 1.72 1.42

Table 3. Isotherm constants of the three-parameter models for Cd (II) and Ni (II) biosorption on FA-treated C. indica biomass at 25˚C

Copyright © 2010. Nova Science Publishers, Incorporated. All rights reserved.

Metal

Equilibrium model

Cadmium

Nickel

Toth qm,T (mg/g) bT (L/g) nT R2 RMSE χ2

2.619 0.101 -0.16 0794 7.339 32.576

0.061 3.671 -0.21 0.999 1.778 1.268

Radke-Prausnitz aR-P (L/g) rR-P (L/mg) βR-P R2 RMSE χ2

-0.273 0.0004 2.55 0.994 2.41 1.75

1.02 0.37 0.99 0.866 3.12 9.62

The Sips isotherms is a combination of the Langmuir and Freundlich isotherm type models and is expected to describe heterogeneous surfaces much better. At low sorbate concentrations it reduces to the Freundlich isotherm, while at high sorbate concentrations it predicts a monolayer adsorption capacity characteristic of the Langmuir isotherm. The model can be written as [58]:

Water Production and Wastewater Treatment, Nova Science Publishers, Incorporated, 2010. ProQuest Ebook Central,

M.M. Montazer-Rahmati, P. Rabbani and A. Abdolali

62 qe =

K s C e βS 1 + aS C e βS

(13)

where KS and aS are the Sips model constants in L/g and L/mg, respectively and βS is the Sips model exponent. The foregoing analysis of isotherm models shows that the best fit for Cd (II) and Ni (II) biosorption is produced by three-parameter isotherm models.

2.8. Effect of Contact Time

Copyright © 2010. Nova Science Publishers, Incorporated. All rights reserved.

As the adsorption process proceeds, the sorbed solute tends to desorb back into the solution. Eventually the rates of adsorption and desorption will attain an equilibrium state. When the system reaches sorption equilibrium, no further net adsorption occurs. The time at which adsorption equilibrium occurs was determined.

Figure 5. Effect of contact time on biosorption of Ni (II) onto C. indica (Ni conc.: 1.0 mmol/L; Temperature: 25˚C; Biomass dosage: 2g/L; pH: 6.0).

The adsorption rate tests were performed on an equilibrium batch basis. 0.06 g/L of the biomass was contacted with a solution bearing a metal concentration of 100 mg/L. The biomass was kept in contact with the metal-bearing solution for different time periods (15, 30, 60 min, 2, 3, 6, 12 and 24 hr). Time zero samples were also taken in these samples for which the biomass was directly separated from the metal-bearing solution within less than one minute of contact time. The very fast sorption and settling of the anaerobic biomass make this material suitable for continuous flow water treatment systems.

Water Production and Wastewater Treatment, Nova Science Publishers, Incorporated, 2010. ProQuest Ebook Central,

Biosorption of Cd (II) and Ni (II) from Aqueous Solutions...

63

Figure 6. Effect of contact time on biosorption of Cd (II) onto C. indica (Cd conc.: 0.25 mmol/L; Temperature: 25˚C; Biomass dosage: 2g/L; pH: 5.5).

From experimental data represented in Fig. 5, the process of biosorption of Ni (II) reached the equilibrium state after approximately 3 h of contact time. This process was rather fast at first and 90% of total biosorption of nickel (II) occurred in the first 60 min. After 1 h, the rate of biosorption was slower and at last no further significant adsorption was noted beyond 3 h. A similar trend for the biosorption of Cd (II) is shown in Fig. 6. It can be seen that equilibrium was achieved within about 3 hours.

Copyright © 2010. Nova Science Publishers, Incorporated. All rights reserved.

2.9. Kinetic Analysis Adsorption kinetics provides valuable information about the mechanism of adsorption and subsequently helps the investigation of the controlling mechanism in the biosorption process which could lie either in the mass transfer or the chemical reaction step [24]. Rate of metal ion removal, which is required for selecting optimum operating conditions for the full scale batch process, can be obtained from a kinetic analysis [61]. Different models can describe biosorption kinetics; two common semi-empirical kinetic models based on adsorption equilibrium capacity are the pseudo-first order model proposed by Lagergren and the pseudo-second order model proposed by Ho and McKay [62].

Figure 7. Linearized pseudo-second-order kinetic model for Cd (II) and Ni (II) ions removal by C. indica biomass at the optimum pH (Cd conc.: 0.25 mmol/L; Ni conc.:1mmol/L; T: 25˚C; Biomass dosage: 2 g/L). Water Production and Wastewater Treatment, Nova Science Publishers, Incorporated, 2010. ProQuest Ebook Central,

M.M. Montazer-Rahmati, P. Rabbani and A. Abdolali

64

Table 4. Comparison between adsorption rate constants, the estimated qe and the coefficients of correlation associated with the pseudo-first-order and the pseudo-secondorder kinetic models at 25˚C

R2

qe (mg/g)

Cadmium

Pseudo-firstorder kinetic model qe K1 (mg/g) 5.111 0.023

0.963

3.80

Pseudo-secondorder kinetic model qe K2 (mg/g) 4.219 3.79E-3

Nickel

5.73

0.830

11.65

11.67

Metal

0.023

Experimental value

4.32E-3

R2 0.986 0.999

In batch systems, adsorption kinetics is most widely described by either the pseudo-firstorder or the pseudo-second-order kinetic models. The linearized pseudo-first-order kinetic model takes the following form [50, 60]: qt = qe − qe exp ( −K 1t )

(14)

where qt and qe are the amounts of metal adsorbed at time t and equilibrium, respectively, and K1 (min-1) is the first-order reaction rate constant. The pseudo-second-order kinetic model considered in this chapter is given as [50, 60]:

Copyright © 2010. Nova Science Publishers, Incorporated. All rights reserved.

t 1 t = + q t K 2qe 2 qe

(15)

where K2 (g mg-1 min-1) is the second-order reaction rate constant. The experimental data presented in Fig. 7 are fitted by the pseudo-second-order kinetic model while the parameters of both models are tabulated in Table 4. It is obvious that the coefficient of correlation (R2: 0.986 and 0.999 for cadmium and nickel biosorption, respectively) for the pseudo-second-order kinetic model is higher in comparison with the pseudo-first-order kinetic model (R2: 0.963 and 0.830) and the calculated value of qe for the pseudo-second-order kinetic model is closer to the experimental value.

2.10. Conclusions The present chapter on biosorption of Cd (II) and Ni (II) from aqueous solutions using brown algae, C. indica, in intact and FA-treated form suggests the following: •

The adsorption of Cd (II) and Ni (II) ions is dependent on pH. Comparison of cadmium (II) and nickel (II) biosorption capacities of brown algae in intact and chemically-modified forms indicates that the maximum sorption capacity of FAtreated C. indica for Cd (II) at an optimum pH of 5.5 is 19.42mg/g and the maximum

Water Production and Wastewater Treatment, Nova Science Publishers, Incorporated, 2010. ProQuest Ebook Central,

Biosorption of Cd (II) and Ni (II) from Aqueous Solutions...

• •

• •

65

Ni (II) biosorption capacity at an optimum pH of 6.0 is 10.06 mg/g (Ci=0.5mmol/L for both Cd (II) and Ni (II) and 2g/L biomass is used in either case). Sorption and desorption cycles have no adverse effect on the biosorption capacity and CaCl2 is a desirable desorbent in this case. Among the two-parameter biosorption isotherms, the Langmuir model better describes the Ni (II) biosorption and the Freundlich model fits the experimental data of Cd (II) biosorption quite well. The maximum monolayer biosorption capacity of FA-treated C. indica for Ni (II) and Cd (II) are 16.17 mg/g and 19.56 mg/g, respectively. Among the three-parameter isotherms, the Toth isotherm best describes the adsorption of Ni (II) and the Radke-Prausnitz model is found to provide the closest fit to the biosorption data of Cd (II). The pseudo-second-order kinetic model fits the experimental data quite well (R2: 0.986 and 0.999 for Cd (II) and Ni (II) biosorption, respectively). It may be concluded that the FA-treated C. indica can be used as a low cost and abundant source for Cd (II) and Ni (II) removal from aqueous solutions.

REFERENCES

Copyright © 2010. Nova Science Publishers, Incorporated. All rights reserved.

[1]

Cruz, C.C.V.; Costa, A.C.A.; Henriques, C.A.; Luna, A.S. Bioresour. Technol. 2004, 91, 249-257. Kinetic modeling and equilibrium studies during cadmium biosorption by dead Sargassum sp. biomass [2] Kaewsarn, P.; Yu, Q. Environ. Pollut., 2001, 112, 209-213. Cadmium (II) removal from aqueous solutions by pre-treated biomass of marine algae Padina sp. [3] Sari, A.; Tuzen, M. J. Hazard. Mater., 2008, 152, 302-308. Biosorption of Pb (II) and Cd (II) from aqueous solution using green alga (Ulva Lactuca) biomass. [4] Lodeiro, P., Cordero, B.; Barriada, J.L.; Herrero, R.; Sastre de Vicente, M.E. Bioresour. Technol., 2005, 96, 1796-1803. Biosorption of cadmium by biomass of brown marine macroalgae. [5] Lodeiro, P., Barriada, J.L.; Herrero, R.; Sastre de Vicente, M.E. Environ. Pollut., 2006, 142, 264-273. The marine macroalga Cysroseira baccata as biosorbent for cadmium (II) and lead (II) removal: Kinetic and equilibrium studies. [6] Esteves, A.J.P.; Valdman, E.; Leite, S.G.F. Biotechnol. Lett., 2000, 22, 499-502. Repeated removal of cadmium and zinc from an industrial effluent by waste biomass Sargassum sp. [7] Kapoor, A.; Viraraghavan, T. Bioresour. Technol., 1995, 53, 195-206. Fungal biosorption–an alternative treatment option for heavy metal bearing wastewaters: a review. [8] Volesky, B. Hydrometallurgy, 2001, 59, 203-216. Detoxification of metal-bearing effluents: biosorption for the next century. [9] Deng, S.; Ting, Y.P. Water Res., 2005, 39, 2167-2177. Characterization of PEImodified biomass and biosorption of Cu(II), Pb(II) and Ni (II). [10] Kurniawan, T. A.; Chana, G. Y. S.; Lo, C. W.; Babel, S. Chem. Eng. J., 2006, 118, 83– 98. Physico–chemical treatment techniques for wastewater laden with heavy metals.

Water Production and Wastewater Treatment, Nova Science Publishers, Incorporated, 2010. ProQuest Ebook Central,

Copyright © 2010. Nova Science Publishers, Incorporated. All rights reserved.

66

M.M. Montazer-Rahmati, P. Rabbani and A. Abdolali

[11] Papadopoulos, A.; Fatta, D.; Parperis, K.; Mentzis, A.; Harambous, K. J.; Loizidou, M. Sep. Purif. Technol., 2004, 39, 181–188. Nickel uptake from a wastewater stream produced in a metal finishing industry by combination of ion-exchange and precipitation methods. [12] Dialynas, E.; Diamadopoulos, E. Desalination, 2009, 238, 302–311. Integration of a membrane bioreactor coupled with reverse osmosis for advanced treatment of municipal wastewater. [13] Ozaki, H.; Sharma, K.; Saktaywin, W. Desalination, 2002, 144, 287–294. Performance of an ultra-low pressure reverse osmosis membrane (ULPROM) for separating heavy metal: effects of interference parameters. [14] Qin, J. J.; Wai, M. N.; Oo, M.H.; Wong, F. S. J. Membr. Sci., 2002, 208, 213–221.A feasibility study on the treatment and recycling of a wastewater from metal plating. [15] Bruggen, B.; Vandecasteele, C. Environ. Pollut. , 2003, 122, 435–445. Removal of pollutants from surface water and groundwater by nanofiltration: overview of possible applications in the drinking water industry. [16] Ahn, K. H.; Song, K.G.; Cha, H. Y.; Yeom, I. T. Desalination, 1999, 122, 77–84. Removal of ions in nickel electroplating rinse water using low-pressure nanofiltration. [17] Qdais, H. A.; Moussa, H. Desalination, 2004, 164, 105–110. Removal of heavy metals from wastewater by membrane processes: a comparative study. [18] Wang, Z.; Liu, G.; Fan, Z.; Yang, X.; Wang, J.; Wang, S. J. Membr. Sci., 2007, 305, 185–195. Experimental study on treatment of electroplating wastewater by nanofiltration. [19] Semerjian, L.; Ayoub, G. M. Adv. Environ. Res., 2003, 7, 389–403. High-pHmagnesium coagulation–flocculation in wastewater treatment. [20] Licsk´o, I. Water Sci. Technol., 1997, 36, 103–110. Realistic coagulation mechanisms in the use of aluminium and iron(III) salts. [21] Ayoub, G. M.; Semerjian, L.; Acra, A.; El Fadel, M.; Koopman, B. J. Environ. Eng., 2001, 127 (3), 196–207. Heavy metal removal by coagulation with seawater liquid bittern. [22] Vik, E. I.; Carlsoon, D. A.; Eikum, A. S.; Gjessing, E. T. Water Res., 1984, 18, 1355– 1360. Electrocoagulation of potable water. [23] Elimelech, M.; O’Melia, C. R. Environ. Sci. Technol., 1990, 24, 1528–1536. Kinetics of deposition of colloidal particles in porous media. [24] Dobrevsky, I.; Todorova-Dimova, M.; Panayotova, T. Desalination, 1996, 108, 277– 280. Electroplating rinse wastewater treatment by ion exchange. [25] Korngold, E.; Belayev, N.; Aaronov, L. Sep. Purif. Technol., 2003, 33 (2), 179–187. Removal of chromates from drinking water by anion exchangers. [26] Ahmed, S.; Chughtai, S.; Keane, M. A. Sep. Purif. Technol., 1998, 13, 57–64. The removal of cadmium and lead from aqueous solution by ion exchange with Na–Y zeolite. [27] Lo, W. H.; Chua, H., Lam, K. H.; Bi, S. P. Chemosphere, 1999, 39 (15), 2723–2736. A comparative investigation on the biosorption of lead by filamentous fungal biomass. [28] Valdimir, S.; Danish, J. M. J. Colloid Interf. Sci., 2002, 250, 213-220. Characterization and metal sorptive properties of oxidized active carbon.

Water Production and Wastewater Treatment, Nova Science Publishers, Incorporated, 2010. ProQuest Ebook Central,

Copyright © 2010. Nova Science Publishers, Incorporated. All rights reserved.

Biosorption of Cd (II) and Ni (II) from Aqueous Solutions...

67

[29] Youssef, A. M.; El-Nabarawy, T. H.; Samra, S. E. Colloids Surf. Eng., 2004, 235, 153160. A Physicochem Sorption properties of chemically-activated carbons 1. Sorption of cadmium(II) ions. [30] Kadirvelu, K.; Thamaraiselvi, K.; Namasivayam, C. Bioresour. Technol., 2001, 76, 6365. Removal of heavy metals from industrial wastewaters by adsorption onto activated carbon prepared from an agricultural solid waste. [31] Ranganathan, K. Bioresour. Technol., 2000, 73, 99-103. Chromium removal by activated carbons prepared from Casurina equisetifolia leaves. [32] Babel, S.; Kurniawan, T. A. J. Hazard. Mater. B, 2003, 97, 219-243. Low-cost adsorbents for heavy metals uptake from contaminated water: a review. [33] Hanif, M. A.; Nadeem, R.; Zafar, M. N.; Akhtar K.; Bhatti, H. N. J. Hazard Mater. 2007, 145, 501–505. Kineticstudies for Ni (II) biosorption from industrial wastewater by Cassia fistula (Golden Shower) biomass. [34] Hawari, A. H.; Mulligan, C. N.; Bioresour. Technol. 2006, 97, 692–700. Biosorption of lead(II), cadmium(II), copper(II) and nickel(II) by anaerobic granular biomass. [35] Naddafi, K.; Nabizadeh, R.; Saeedi, R.; Mahvi, A.H.; Vaezi, F.; Yaghmaeian, K.; Ghasri, A.; Nazmara, S. J. Hazard. Mater. 2007, 147, 785-791. Biosorption of lead(II) and cadmium(II) by protonated Sargassum glaucescens biomass in a continuous packed bed column. [36] Williams, C. J.; Aderhold, D.; Edyvean, R. g. J. J. Water Res. 1998, 32, 216-224. Comparison between biosorbents for the removal of metal ions from aqueous solution. [37] Romera, E.; Gonzalez, F.; Balleter, A.; Blazquez, M. L.; Munoz, J. A. J. Bioresour. Technol. 2008, 99, 4684-4693. Biosorption of heavy metals by Fucus spiralis. [38] Ricordel, S.; Taha, S.; Cisse, I.; Dorange, G. J. Sep. Purif. Technol. 2001, 24, 389-401. Heavy metals removal by adsorption onto peanut husks carbon: characterization, kinetic study and modeling. [39] Pejic, B.; Vukcevic, M.; Kostic, M.; Skundric, P.; J. Hazard. Mater. 2009, 164, 146– 153. Biosorption of heavy metal ions from aqueous solutions by short hemp fibers: Effect of chemical composition. [40] Kapoor, A.; Viraraghavan, T. Bioresour. Technol. 1998, 63, 109-113. Biosorption of heavy metals on aspergillus niger: effect of pretreatment. [41] Davis, T. A.; Volesky, B. J. Water Res. 2005, 39, 239-247. Biosorption of La, Eu and Yb using Sargassum biomass. [42] Davis, T. A.; Volesky, B.; Mucci, A. J. Water Res. 2003, 37, 4311-4330. A review of the biochemistry of heavy metal biosorption by brown algae. [43] Hashim, M.; Chu, K. H. J. Chem. Eng. 2004, 97, 249-255. Biosorption of cadmium by brown, green and red seaweeds. [44] Tsui, M. T. K.; Cheung, K. C.; Tam, N. F. Y.; Wong, M. H. J. Chemosphere. 2006, 65, 51-57. A comparative study on metal sorption by brown seaweed. [45] Vijayaraghavan, K.; Palanivelu, K.; Velan, M. J. Sep. Purif. Technol. 2005, 44, 53-59. Biosorption of cobalt(II) and nickel(II) by seaweeds: batch and column studies. [46] Bold, H. C.; Wynne, M. J. Introduction to the Algae Structure and Reproduction; Second Edition; Prentice-Hall, Inc.: Englewood Cliffs, New Jersey 1985; Vol. 1, pp 516-525. [47] Kalyani, S.; Srinivasa Rao, P.; Krishnaiah, A. J. Chemosphere. 2004, 57, 1225-1229. Removal of nickel(II) from aqueous solution using marine algae as sorbing biomass.

Water Production and Wastewater Treatment, Nova Science Publishers, Incorporated, 2010. ProQuest Ebook Central,

Copyright © 2010. Nova Science Publishers, Incorporated. All rights reserved.

68

M.M. Montazer-Rahmati, P. Rabbani and A. Abdolali

[48] Khani, M. H.; Keshtkar, A. R.; Ghandchi, M.; Pahlavanzadeh, H. J. Hazard. Mater. 2008, 150, 612-618. Equilibrium, kinetic and thermodynamic study of the biosorption of uranium onto Cystoseira Indica algae. [49] Schiewer, S.; Wong, M. H. J. Chemosphere. 2000, 41, 271-282. Ion strength effect in biosorption of metals by marine algae. [50] Chen, Z.; Ma, W.; Han, M. J. Hazard. Mater. 2008, 155, 327-333. Biosorption of nickel and copper onto treated alga (Undria Pinnatifida): Application of isotherm and kinetics study. [51] Yang, L.; Chen, P. J. Bioresour. Technol. 2008, 99, 297-307. Biosorption of hexavalent chromium onto intact and chemically modified Sargassum sp. [52] Matheickal, J.T.; Yu, Q. J. Bioresour. Technol. 1999, 69, 223-229. Biosorption of lead(II) and copper(II) from aqueous solutions by pre-treated biomass of Australian marine algae. [53] Leusch, A.; Holan, Z. R.; Volesky, B. J. Chem. Tech. Biotechnol. 1995, 62, 279-288. Biosorption of heavy metals (Cd, Cu, Ni, Pb, Zn) by chemically rein-forced biomass of marine algae. [54] Luo, F.; Liu, Y. H.; Li, X. M.; Xuan, Z. X.; Ma, J. T. J. Chemosphere. 2006, 64, 1122– 1127. Biosorption of lead ion by chemicallymodified biomass of marine brown algae Laminaria japonica. [55] Akar, T; Kaynak, Z.; Ulusoy, S; Yuvaci, D.; Ozsari, G.; Akar, S. T.; J.Hazard. Mater. 2009, 163, 1134–1141. Enhanced biosorption of nickel(II) ions by silica-gelimmobilized waste biomass: Biosorption characteristics in batch and dynamic flow mode. [56] Schiewer, S.; Volesky, B. J. Environ. Sci. Technol. 1995, 29, 3049– 3058. Modeling of the proton-metal ion exchange in biosorption. [57] Gerringa, L. J. A.; Herman, P. M. J.; Poortvliet, T. C. W. J. Mar. Chem. 1995, 48, 131– 142. Comparison of the linear Van den Berg/Ruzic transformation and a non-linear fit of the Langmuir isotherm applied to Cu speciation data in the estuarine environment. [58] Vijayaraghavan, K.; Padmesh, T. V. N.; Palanivelu, K.; Velan, M. J. Hazard. Mater. 2006, B133, 304-308. Biosorption of nickel(II) ions onto Sargassum wightii: Application of two-parameter and three- parameter isotherm models. [59] Daneshvar, N.; Salari, D.; Aber, S. J. Hazard. Mater. 2002, 94, 49–61. Chromium adsorption and Cr(VI) reduction to trivalent chromium in aqueous solutions by soya cake. [60] Basha, S.; Murthy, Z. V. P. J. Process Biochem. 2007, 42, 1521-1529. Kinetic and equilibrium models for biosorption of Cr(VI) on chemically modified seaweed, Cystoseira indica. [61] Alyüz, B.; Veli, S. J. Hazard. Mater. 2009, 167, 482-488. Kinetics and equilibrium studies for the removal of nickel and zinc from aqueous solutions by ion exchange resin. [62] Ho, Y. S.; McKay, G. J. Proc. Biochem. 1999, 34, 451–459. Pseudo-second order model for sorption processes.

Water Production and Wastewater Treatment, Nova Science Publishers, Incorporated, 2010. ProQuest Ebook Central,

In: Water Production and Wastewater Treatment Editors: B. Antizar-Ladislao and R. Sheikholeslami

ISBN 978-1-61728-503-5 © 2011 Nova Science Publishers, Inc.

Chapter 5

MULTICOMPONENT REMOVAL OF HEAVY METALS FROM AQUEOUS SOLUTION USING LOW-COST SORBENTS Hilda Elizabeth Reynel-Avila, Didilia Ileana Mendoza-Castillo, Virginia Hernández-Montoya, and Adrián Bonilla-Petriciolet *1 Instituto Tecnológico de Aguascalientes, Aguascalientes, México, 20256

Copyright © 2010. Nova Science Publishers, Incorporated. All rights reserved.

ABSTRACT Recently, some materials from natural or industrial origin have been received attention as possible sorbents for the removal of heavy metals from water due to their acceptable sorption behavior, low cost and large availability. Most of the heavy metals sorption studies using low-cost sorbents are concerned with monometallic systems. However, the industrial wastewaters usually contain more than one metallic specie and, therefore, it is necessary to study and understand the sorption behavior under competitive conditions (i.e., when several metallic species are present). This chapter provides a brief overview of various low-cost sorbents that have been used as sorbent for heavy metal removal in aqueous solution under competitive conditions. The discussion is especially focused on the results reported for sorption studies using some sorbents from different origin, their relative advantages and sorption capacities in multimetallic solutions. Finally, the description and the application of classical multicomponent sorption models are provided including some guidelines for a reliable parameter estimation procedure.

1

* Corresponding author: [email protected]

Water Production and Wastewater Treatment, Nova Science Publishers, Incorporated, 2010. ProQuest Ebook Central,

70

H. E. Reynel-Avila, D. I. Mendoza-Castillo, V. Hernández-Montoya et al.

Copyright © 2010. Nova Science Publishers, Incorporated. All rights reserved.

1. INTRODUCTION Industrial wastewaters are important source of environmental pollution due to the high content of heavy metal ions. These metals are toxic, non-biodegradable, can be incorporated relatively easy in the food chain, and tend to accumulate causing several diseases and health disorders in humans, and other living organisms [1,2]. In particular, cadmium (Cd+2), lead (Pb+2), nickel (Ni+2), mercury (Hg+2), zinc (Zn+2), copper (Cu+2) and chromium (Cr+3) are the most common metals found in effluents of a large number of industries. The toxicity of these heavy metals has been well documented in literature and their presence in water resources and wastewaters is a potential risk for the environment and public health [1-4]. Thus, it is necessary to design technically and economically feasible processes to minimize the heavy metal pollution and to reduce the risk associated to it. To date, several studies have focused on the development of effective and low-cost strategies for heavy metal removal from water. The conventional approaches include: ionexchange, reverse osmosis, evaporation, chemical precipitation, electrochemical treatment, coagulation, solvent extraction, chemical oxidation-reduction, and sorption [1,3,5]. Unfortunately, some methods show technical limitations, are expensive and ineffective when the metal concentrations are in the range of 1 – 100 mg/L. Besides, available technologies may show high sensitivity to operational conditions, high energy consumption, and also produce large quantity of residual sludge [3,5]. The sorption process is one of the most important methods for treatment of wastewaters polluted by heavy metal ions. This technology offers several advantages for facing water pollution and is even better than other techniques due to its effectiveness, feasibility, versatility, simplicity of design, easiness of operation, possibility of metal recovery, and low cost if a proper sorbent is used [6,7]. Activated carbon is the most known sorbent and has been used for a long time to remove various pollutants including heavy metal ions [4,7,8]. However, the synthesis and regeneration cost of commercial carbons have encouraged the application of low-cost renewable materials. In this context, natural materials, biomasses and wastes from industrial or agricultural operations that are available in large quantities and require little processing, may be used as low-cost sorbents for improving water quality [1,4,5,7]. These sorbents, which can be used either directly or after a suitable pre-treatment, may show competitive sorption capacities and offer more advantages than commercial activated carbons [1,4,5]. In fact, its use in water treatment has become a current tendency in sorption research and is an attractive approach for the field of environmental engineering. Reviews of the feasibility and application of a wide variety of low-cost sorbents for the treatment of polluted water by heavy metal ions can be found in Bailey et al. [1], Wan Ngah and Hanafiah [2], Sud et al. [4], Wang and Chen [5], Demirbas [7] and Babel and Kurniawan [8]. These sorbents include ashes, zeolites, clays, biomasses, industrial wastes, and others. In the majority of these works the sorption of a single metal ion has been studied principally and a limited emphasis has been given to the study of multicomponent systems. However, the industrial effluents contain several metallic species and, as a consequence, it is necessary to evaluate the behavior of available sorbents under these conditions. Note that the simultaneous removal of heavy metal ions from solutions containing two or more species plays an important role for the design and operation of water purification processes, because

Water Production and Wastewater Treatment, Nova Science Publishers, Incorporated, 2010. ProQuest Ebook Central,

Multicomponent Removal of Heavy Metals…

71

multicomponent sorption studies are useful to identify the competitive effects on sorbent behavior between several metal species. When two or more metal ions are present in solution, they may increase, decrease or may not change the metal-ion sorption capacity of the sorbent [9,10]. Herein, it is convenient to remark that the study, analysis and the interpretation of multicomponent sorption process have also proved to be complex because sorption process is affected by several factors and there are various mechanisms involved in the removal of metal ions, which may occur simultaneously [11]. Actually, experimental sorption data on multimetallic systems are very limited and further studies are necessary to improve the knowledge of multicomponent sorption for the removal of hazardous metallic species using low-cost sorbents. This chapter provides a brief overview of low-cost sorbents for multicomponent heavy metal removal in aqueous solution. The sorption behavior of various sorbents, including industrial by-products, wastes and natural materials, are discussed and analyzed. Finally, the application of multicomponent sorption models for data correlation is described, and the principal aspects for a reliable modeling procedure are highlighted.

Copyright © 2010. Nova Science Publishers, Incorporated. All rights reserved.

2. DESCRIPTION OF LOW COST SORBENTS USED IN THE REMOVAL OF HEAVY METALS FROM MULTICOMPONENT MIXTURES In contrast to single-metal sorption studies, a limited number of low-cost sorbents has been studied for heavy metal removal under competitive conditions in multimetallic aqueous systems. Tables 1 and 2 present a summary of the most representative low-cost sorbents used to remove heavy metal ions in multicomponent systems. These tables show the studied metallic species and the sorption experimental conditions. The sorbents include natural polymers, incineration wastes, materials from biological origin, minerals and sediments, industrial by-products and wastes. Note that this chapter does not cover the use of biological sorbents such as bacteria, yeasts and fungi under competitive conditions. For interested readers, an excellent review of the application of several biosorbents for removal of heavy metals from aqueous solutions has been reported by Wang and Chen [5]. Basically, a multicomponent solution may exhibit three possible types of sorption effects under competitive conditions: a) synergism: the sorption of a sorbate increases when there are other sorbates in the mixture, b) antagonism: the sorption of a sorbate decreases when there are other sorbates in the mixture, and c) non-interaction: the mixture has no effect on the sorption of each sorbate in the mixture. In literature, the analysis of competitive effects is based on the ratio of the sorption capacity of one metal ion in the presence of the other metal ions in multicomponent solution, qei,mix, to the sorption capacity of same metal ion when it is present alone in the solution, qei,0 [12-14]. Using this approach, when qei,mix / qei,0 > 1 sorption is promoted by the presence of other metal ions (synergism); if qei,mix / qei,0 = 1 the metals have no effect on each other (non-interaction); and if qei,mix / qei,0 < 1 sorption is suppressed by other metal ions (antagonism). Typically, this analysis is applied for characterizing and studying the sorbent behavior under competitive conditions. With illustrative purposes, a set of sorption studies of various low-cost sorbents are discussed in the following. This discussion comprises the sorbent description and its behavior

Water Production and Wastewater Treatment, Nova Science Publishers, Incorporated, 2010. ProQuest Ebook Central,

72

H. E. Reynel-Avila, D. I. Mendoza-Castillo, V. Hernández-Montoya et al.

and competitive effects in multimetallic solutions. The results of data modeling and some aspects about mechanisms involved in the removal of heavy metal ions are also considered. Finally, the sorption capacities of these sorbents under competitive conditions will be compared and discussed in the section 3.

2.1. Alginate Alginate is a polysaccharide distributed widely in the cell walls of brown algae. It has been extensively used as immobilization material for various applications, and few studies have tested its behavior as a heavy metal sorbent. Specifically, calcium alginate beads obtained from brown algae have been used for the simultaneous removal of Cu+2, Cd+2 and Pb+2 ions in binary solutions [15,16]. Sorption experiments were performed at pH 4.5 and 25 °C with metal concentrations up to 10 mmol/L. The sorbent dosage was 1 mg/mL and the particle size was 0.7 mm. Experimental data revealed that Cu+2 and Pb+2 ions are sorbed preferably more than Cd+2 ions. These results suggest antagonistic competitive effects especially for Cd+2 ions. In particular, the sorption behavior of alginate for Pb+2 ions is less sensitive to the presence of both Cu+2 and Cd+2 ions in solution. In this study, Langmuir and Sips multicomponent models were used for data regression and the results indicated that the non-modified Langmuir model provided the best correlation of experimental data. Papageorgiou et al. [16] suggested that the main mechanism involved in the sorption of heavy metal ions using alginate is considered to be ion exchange.

Copyright © 2010. Nova Science Publishers, Incorporated. All rights reserved.

2.2. Ashes Incineration wastes from natural or anthropogenic sources have good sorption properties and have been used for multicomponent removal of heavy metals [10,14,17-19]. These wastes include rice husk ash, volcanic ash, bagasse fly ash, and others (see Tables 1 and 2). Specifically, Srivastava et al. [14] reported the application of bagasse fly ash for the simultaneous removal of Cd+2 and Ni+2 ions from water. Bagasse fly ash is a waste of sugar industry and is obtained from bagasse-fired boilers. The results of X-ray spectrum of this waste indicated the presence of alumina, silica, calcium oxide, calcium metasilicate, and calcium silicate while FTIR analysis suggested the presence of silanol, aldehydes, ketones, lactones and carboxyl-carbonate groups, which are useful for heavy metal removal. This sorbent was used without treatment in sorption experiments and its BET surface area was ≅ 169 m2/g. Multimetallic removal studies were carried out at 30 °C and pH 6 using metal concentrations from 0.1 to 1.7 mmol/L and a sorbent dosage of 10 mg/mL. Results indicated that the equilibrium uptake of one ion (Cd+2 or Ni+2) decreased with increasing concentrations of other ion. For all tested conditions, there is an antagonistic sorption effect between Cd+2 and Ni+2 ions. Comparing the sorbent behavior, these studies revealed that the preference for Ni+2 is greater than Cd+2. The sorption data were fitted to several models including Langmuir, Freundlich and Redlich-Peterson multicomponent isotherms. In particular, the extended Freundlich model provided the best correlation of experimental data in comparison to the

Water Production and Wastewater Treatment, Nova Science Publishers, Incorporated, 2010. ProQuest Ebook Central,

Multicomponent Removal of Heavy Metals…

73

other models. These authors concluded that as the solution pH increases, the metal sorption improves due to electrostatic attraction. Table 1. Summary of sorption studies in binary systems for heavy metal removal from aqueous solution using low-cost sorbents Sorbent Alginate Bagasse fly ash Bone char Carbonized sewage sludge Chicken feathers Chitosan Crab shell Iron oxide-coated sediment Lignin Olive pomace Olive mill waste Olive stone waste Orange wastes Pine bark Peat

Rice husk ash

Copyright © 2010. Nova Science Publishers, Incorporated. All rights reserved.

1

Metal ions studied +2

+2

+2

+2

+2

+2

Cd - Cu , Pb - Cd , Cu - Pb Cd+2 - Ni+2 Cd+2 - Cu+2, Cd+2 - Zn+2, Cu+2 - Zn+2 Hg+2 - Pb+2, Cu+2 - Hg+2, Cr+3 - Hg+2, Cu+2 - Pb+2, Cu+2 - Cr+3, Pb+2 - Cr+3 Cu+2 - Zn+2, Zn+2 - Ni+2, Cu+2 - Ni+2 Cu+2 - Zn+2, Zn+2 - Ni+2, Cu+2 - Ni+2, Cu+2 - Hg+2 Pb+2 - Cd+2, Cd+2 - Cr+3, Pb+2 - Cr+3 Cu+2 - Ni+2 Cu+2 - Zn+2, Cd+2 - Cu+2, Cd+2 - Zn+2 Cd+2 - Cu+2 Pb+2 - Cd+2, Cu+2 - Pb+2, Cd+2 - Cu+2 Cd+2 - Cu+2, Cu+2 - Ni+2, Cu+2 - Pb+2, Cd+2 - Ni+2, Pb+2 Ni+2, Pb+2 - Cd+2 Cd+2 - Zn+2, Pb+2 - Cd+2, Pb+2 - Zn+2 Cd+2 - Cu+2, Cu+2 - Ni+2, Cu+2 - Pb+2, Cd+2 - Ni+2, Pb+2 Cd+2, Pb+2 - Ni+2 Cd+2 - Cu+2, Cd+2 - Zn+2, Cu+2 - Zn+2, Cu+2 - Ni+2, Cu+2 Cr+3, Cd+2 - Cr+3, Cu+2 - Pb+2, Pb+2 - Cd+2, Pb+2 - Zn+2 Cd+2 - Zn+2, Cd+2 - Ni+2

Conditions 1 T, °C 25 30 20 25

pH 4.5 6 4.9 3-5

15,16 14 20,22,23 46

25 20 - 25

NS 2-5

24 27,28

30 20

5 5

30 33

25 25 20 20

4.5 4-5 7 5.5

12 34 35 36

NS NS

4 4

37 45

20 - 25

4-6

38-42,44

30

6

17,18

Ref.

NS indicates that the pH or temperature is not specified in the reference.

Table 2. Summary of sorption studies in ternary, quaternary and quinary systems for the removal of heavy metals in aqueous solution using low-cost sorbents Sorbent Bone char Chicken feathers

Metal ions studied +2

+2

+2

Conditions 1 T, °C 20 25 - 45

Zn - Cd - Cu Cu+2 - Pb+2 - Hg+2 Cu+2 - Pb+2 - Zn+2 - Cd+2 - Ni+2 30 Crab shell Pb+2 - Cd+2 - Cr+3 NS Clinoptilolite Cu+2 - Cd+2 - Ni+2 - Pb+2 25 Lignin Zn+2 - Cd+2 - Cu+2 30 Low-grade phosphate rock Cu+2 - Zn+2 - Pb+2 20 Olive mill waste Cu+2 - Cd+2 - Pb+2 Peat Zn+2 - Cd+2 - Cu+2 20 - 25 Cu+2 - Cd+2 - Pb+2 Pb+2 - Zn+2 - Cd+2 NS Pine bark Cu+2 - Cd+2 - Ni+2 Cu+2 - Cd+2 - Pb+2 Cd+2 - Ni+2 - Pb+2 Ni+2 - Cu+2 - Pb+2 Cu+2 - Cd+2 - Ni+2 - Pb+2 Rice husk ash Cd+2 - Ni+2 - Zn+2 30 20 Volcanic lava ash Cd+2 - Cu+2 - Cr+3 1 NS indicates that the pH or temperature is not specified in the reference.

Water Production and Wastewater Treatment, Nova Science Publishers, Incorporated, 2010. ProQuest Ebook Central,

pH 4.9 1.9 - 5.6

Ref. 23 25

5 6.2 4.5 NS 7 4.5 - 6

30 31 12 32 35 13,39

4

45

6 6

10 19

Copyright © 2010. Nova Science Publishers, Incorporated. All rights reserved.

74

H. E. Reynel-Avila, D. I. Mendoza-Castillo, V. Hernández-Montoya et al.

On the other hand, rice husk ash is an agricultural waste obtained during the burning of rice husk. This low-cost sorbent has been tested in both binary and ternary solutions of heavy metals. For example, Srivastava et al. [10,17,18] reported the simultaneous removal of Cd+2, Ni+2 and Zn+2 from binary and ternary mixtures using rice husk ash. This sorbent was obtained from a paper mill industry and was used in sorption experiments without treatment. This waste has a surface area of 36.44 m2/g and contains several functional groups as -CO-, OH, -Si-OH-, -SiH, -C-OH- that may interact with heavy metals. Multicomponent sorption experiments were performed at 30 °C and pH 6 using different metal concentrations (0.1 – 1.7 mmol/L). Equilibrium studies indicated that the combined effect of both binary and ternary mixtures on sorption behavior is of antagonism, where the equilibrium uptake of each ion decreases with increasing concentration of other ions. In general, for binary systems, the sorption capacity of rice husk ash for Cd+2 is less than Ni+2, and the sorbed amount of Zn+2 is larger than Cd+2. It appears that, under competitive conditions, the Ni+2 sorption was less affected by the presence of Cd+2 ions in comparison to the inhibition exerted in the reverse situation; while the sorption capacity of rice husk ash for Cd+2 is less than Zn+2. In ternary mixture, the affinity of rice husk ash for Zn+2 is higher than the Ni+2 and Cd+2 ions. Several non-modified and modified multicomponent models were used for data correlation. The extended Freundlich model offers the best fits for both sets of binary experimental data; while Sheindorf-Rebuhn-Sheintuch model provided the best regression of ternary experimental sorption data in comparison to the other models. Finally, it appears that the removal of these metal ions using rice husk ash is a combination of two mechanisms: chemisorption and electrostatic attractions. The volcanic lava ash has been also studied as an alternative low-cost sorbent for heavy metal removal from wastewaters [19]. This siliceous material is released into the atmosphere from volcanoes in quantity which is estimated up to 150 million tons per year. The sorption capabilities of this waste were tested in multimetallic experiments using a mixed solution of three ions: Cd+2, Cu+2 and Cr+3. Sorption experiments were performed at pH 6 and 20 °C employing different equimolar concentrations of each metal (0.025 – 0.1 mmol/L). In this study, Cu+2 ions were sorbed preferably more than Cd+2 and Cr+3 ions at tested conditions. Authors commented that at low initial metal concentration, the sorption behavior was found comparable with those from single-metal tests (i.e., under these conditions, no competitive effects were found). However, the multicomponent isotherms were not performed and, as a consequence, the maximum sorption capacities for competitive conditions are unknown. According to author’s discussion, the lava ash is a material of complex structure and heterogeneous composition, where several mechanisms may act in the removal of heavy metal.

2.3. Bone Char Recently, the interest in the application of materials from biological origin (plant or animal) in heavy metal removal has increased. In particular, animal bones are composed in 65 – 70% of inorganic material, mainly hydroxyapatite. This hydroxyapatite in bones is useful for sorbing organic and inorganic species from aqueous solutions, including heavy metal ions. Bone char is a heterogeneous sorbent which is derived from the pyrolisis of animal bones. It mainly contains carbon and hydroxyapatite randomly distributed on the surface [20].

Water Production and Wastewater Treatment, Nova Science Publishers, Incorporated, 2010. ProQuest Ebook Central,

Multicomponent Removal of Heavy Metals…

75

Traditionally, this sorbent has been used to decolorize sugar solutions in the sugar industry. But, recent studies have shown that bone char has significant sorption capacities for various metal ions [20,21]. However, few studies have been performed under competitive conditions [21-23]. Reported studies have tested the sorption behavior of bone char using binary and ternary solutions of Cu+2, Cd+2 and Zn+2 ions [20-23]. Several kinetic and equilibrium sorption experiments were carried out by Cheung et al. [22] and Cheung et al. [23] to assess the influence of co-ion and its concentration on the metal uptake of bone char at pH 4.9 and 20 °C using metal concentrations from 0.1 – 6 mmol/L. These studies revealed an antagonistic effect of Cu+2, Cd+2 and Zn+2 ions in both binary and ternary solutions. The sorption of Cd+2 and Zn+2 ions was negatively impacted by Cu+2 ions, especially at high concentrations of the systems. Cd+2 or Zn+2 ions were competing with and being displaced by the Cu+2 ions in the multicomponent systems. In binary solution Cd+2-Zn+2, the uptake was higher for Cd+2 ions. It appears that the presence of Cd+2 has a more competitive effect on Cu+2 sorption than Zn+2. However, the sorption capacities for the Cu+2 were higher than the Cd+2 and Zn+2, and the sorption capacity for Cd+2 was also higher than Zn+2. Overall, the sorbed amount of all metal ions in binary systems was higher than those reported for ternary solutions. Mass transfer equations were used to model the multicomponent sorption behavior in these binary and ternary systems obtaining satisfactory fits [20-23]. With respect to the removal mechanisms, these studies proposed that the sorption of metal ions on bone char may involve ion exchange, surface sorption and chelation.

Copyright © 2010. Nova Science Publishers, Incorporated. All rights reserved.

2.4. Chicken Feathers The fibrous proteins contained in some animal processing wastes are abundant resources to obtain potential sorbents for heavy metal removal from water. The chemical composition of these proteins involves carboxyl, hydroxyl and amine-groups that may act as binders for metal ions. In particular, the keratin is a protein that contains functional groups useful for the sorption of metal species [24-26]. Chicken feathers are an important source of this protein because they consist of about 91% keratin. This feather keratin is a by-product in the poultry industry and is available in significant amounts because a chicken has about 5 to 7% of its body weight in feathers. Sorption processes using chicken feathers have proven to be effective and economically feasible to remove heavy metals from water. However, most of the studies on the removal of heavy metals ions using chicken feathers have been focused on the uptake of single metals, and few studies have reported the removal of heavy metals in multicomponent systems [24,25]. Details of multicomponent sorption studies using chicken feathers are given in Tables 1 and 2. For example, Al-Asheh et al. [24] studied the removal of binary systems composed of +2 Cu , Zn+2 and Ni+2 ions using metal concentrations from 0.3 to 1.7 mmol/L at 25 °C. Chicken feathers were used in sorption experiments without treatment. This study indicated that the sorption of all metal ions was suppressed by the presence of the other ion in all binary systems tested, and the level of suppression in the metal sorption increased as the concentration of the other metal ion in the system increased. The removal of Zn+2 and Cu+2 ions was significantly affected by the presence of Ni+2 ion especially at high metal concentrations in both Zn+2-Ni+2 and Cu+2-Ni+2 solutions, while the presence of Cu+2 ions appears to have a significant effect on Zn+2 sorption. Authors concluded that the competitive

Water Production and Wastewater Treatment, Nova Science Publishers, Incorporated, 2010. ProQuest Ebook Central,

76

H. E. Reynel-Avila, D. I. Mendoza-Castillo, V. Hernández-Montoya et al.

removal of these metal ions may be attributed to the electronegativity of each metallic species. Finally, Langmuir, Freundlich and Sips multicomponent equations were used for data correlation employing the parameters obtained from single-solute systems. Freundlich model provided the better fit for all sorption data, while Langmuir model showed the worst performance. In other study, Kar and Misra [25] reported the removal of Cu+2, Pb+2 and Hg+2 in a ternary metal solution containing 2 mg/L of each metal at three pH values: 1.9, 4.5 and 5.6. The sorbent behavior was tested using metal removal percentages and authors reported that the removal percentage was of the following order: Pb+2 > Hg+2 > Cu+2. It appears that the sorption of Pb+2 or Cu+2 onto chicken feathers is better since the atomic weight of these metal ions is higher. For the case of Hg+2, they suggested that this metal may form an anionic complex in solution and thus chicken feathers showed better sorption behavior at lower pH, unlike Pb+2 and Cu+2. In addition, metal removal studies were carried out using a multicomponent solution with an initial concentration of 2 mg/L of Cu+2, Pb+2, Zn+2, Cd+2 and Ni+2 at three different pH values: 4.2, 5, and 5.6. The highest metal removal was obtained at pH 5.6 and the removal percentage was: Pb+2 > Cu+2 > Cd+2 > Zn+2 > Ni+2. Kar and Misra [25] noted that chicken feathers removed very low amounts of Zn+2, Cd+2 and Ni+2 under these conditions. Unfortunately, these authors do not report the multicomponent isotherms and as a consequence, the maximum sorption capacities under competitive conditions are unknown. This study suggested that the metal removal by keratin may occur by a combination of both physisorption and chemisorption.

Copyright © 2010. Nova Science Publishers, Incorporated. All rights reserved.

2.5. Chitosan Chitosan is a partially acetylated glucosamine polymer and it is produced cheaply from deacetylation of chitin, which is one of the most abundant polymers in the nature and is the major component of crustacean shells that are wastes of seafood processing industries [2729]. Several authors have found that chitosan has unique sorption and chelating properties for a number of heavy metal ions. In fact, the removal of various metallic species has been studied extensively using raw and chemically modified chitosan, principally in single-metal systems [1]. For the case of competitive removal of heavy metals, Juang and Shao [27] reported the sorption of Cu+2, Ni+2 and Zn+2 ions in binary solutions using glutaraldehyde cross-linked chitosan beads obtained from lobster shell wastes. This sorbent has a surface area of 60 m2/g. Experiments were carried out at 25 °C and different pH conditions using initial metal concentrations of 0.77 – 17 mmol/L and a sorbent dosage of 1 mg/mL. This study revealed a preferential sorption of Cu+2 ions on chitosan beads and an antagonistic competitive effect was determined for these binary systems. Cu+2 sorption appears to be preferred in the binary systems Cu+2-Ni+2 and Cu+2-Zn+2, while the removal of Zn+2 is favored in the binary solution of Ni+2-Zn+2. A simplified model was developed for sorption data regression by considering possible competitive reactions, including the protons in the solutions. The amino and hydroxyl groups on chitosan were identified as active sites for heavy metal removal. In other study, Vieira et al. [28] investigated the binary sorption of Cu+2 and Hg+2 metal ions using natural and crosslinked chitosan membranes. The influence of metal concentration and the crosslinking agent in the mixture (i.e., glutaraldehyde and epichlorohydrin) were

Water Production and Wastewater Treatment, Nova Science Publishers, Incorporated, 2010. ProQuest Ebook Central,

Multicomponent Removal of Heavy Metals…

77

evaluated. Porous membranes of chitosan were prepared and the raw chitosan membranes were crosslinked using aqueous glutaraldehyde and epichlorohydrin solutions. Batch sorption experiments were conducted using a sorbent dosage of 12 mg/mL at pH 5 and 20 °C. For natural chitosan membranes, the Hg+2 ions were more sorbed than the Cu+2 ions, indicating a stronger interaction between the first metallic specie and the sorbent. The sorbed amount of Cu+2 ions in binary solutions was significantly reduced in comparison with the single system, which was attributed to the competition of Hg+2 ions. On the other hand, the results for glutaraldehyde-crosslinked and epichlorohydrin-crosslinked chitosan membranes indicated that, in some cases, the sorbed amount in binary systems was higher than those reported for monometallic systems, suggesting a synergy of electrostatic and chelation mechanisms. Different versions of Langmuir-based model were used for data correlation obtaining a well fit of experimental data. Literature suggested that amino and hydroxyl groups of chitosan may participate in sorption of heavy metals [27,29]. It appears that the interaction of these groups with metallic ions can depend of the pH and also of the main solution components. Finally, the sorption behavior of chitosan is also related to its deacetylation degree where the sorption capacity improves as the deacetylation degree increases [29].

Copyright © 2010. Nova Science Publishers, Incorporated. All rights reserved.

2.6. Crab Shell Recently, Kim [30] reported the application of crab shells as sorbent for multicomponent heavy metal removal. Most crab products are used for food processing and the crab shells can be obtained cheaply from process wastes. Shell of Chinonecetes was obtained as waste from a crabmeat processing plant. A particle size of 420 – 841 μm with a specific surface area of 13.35 m2/g was considered for the sorption experiments. The chemical composition analysis indicated that this waste contained protein, ash, lipid, and chitin. Experimental tests were done with binary and ternary mixtures of Cd+2, Pb+2 and Cr+3 using a sorbent dosage of 1 mg/mL and an equimolar concentration of 0.5 mmol/L for each metallic specie at pH 5 and 30 °C. Sorption data suggested that the presence of Cd+2 did not significantly affect the removal of both Pb+2 and Cr+3 in both binary and ternary solutions. It appears that Cr+3 and Pb+2 ions had severe inhibition effect on the removal of Cd+2 especially in ternary mixtures. This study concluded that Pb+2 and Cr+3 can be easily removed in mixed metal ions systems using this sorbent. Finally, the data modeling was not performed in this study.

2.7. Natural Minerals and Sediments In recent years, the study of different types of natural low-cost sorbents such as minerals and sediments has received more attention. These natural sorbents are abundant in nature and are considered suitable for the removal of trace heavy metals. In this context, natural zeolites have gained a significant interest in sorption field due to their valuable sorption characteristics provided by combination of ion-exchange and molecular-sieve properties, their nontoxic nature, and wide availability [31]. Clinoptilolite is one of the most common natural zeolites and its large industrial deposits are connected with volcanic sedimentary high-silica rocks. Several studies have investigated various aspects of heavy metal removal from aqueous

Water Production and Wastewater Treatment, Nova Science Publishers, Incorporated, 2010. ProQuest Ebook Central,

Copyright © 2010. Nova Science Publishers, Incorporated. All rights reserved.

78

H. E. Reynel-Avila, D. I. Mendoza-Castillo, V. Hernández-Montoya et al.

solutions by clinoptilolite, including the chemical modification of this zeolite to improve its sorption behavior. However, in few works the influence of other ions on heavy metal removal from mixed solutions has been studied. For instance, the sorption behavior of clinoptilolite rock was tested in multicomponent solutions of Ni+2, Cu+2, Pb+2 and Cd+2 ions [31]. These studies were performed at pH 6.2 using an initial metal concentration of 20 mg/L for each metallic ion. For this zeolite, Ni+2 sorption decreased significantly due to competition with other ions. Unfortunately, multicomponent isotherms were not performed and the competitive effects of all metal ions can not be clearly identified. It is convenient to mention that the sorption behavior of zeolites is affected by several factors such as metal concentration, temperature, pH, and crystalline structure of the zeolites. Alternatively, other low-cost mineral sorbents have been successfully utilized for the removal of heavy metal ions from water. For example, Prasad et al. [32] reported the application of low-grade phosphate rock on multicomponent sorption of Pb+2, Cu+2 and Zn+2 ions. This mineral is considered as waste for fertilizer industries due to its very low phosphate value and its use in sorption process is a viable alternative for its disposal. Samples of lowgrade rock phosphate from India were collected. X-ray diffraction analysis of this mineral indicated the presence of calcite, fluorapatite, quartz, dolomite and iron oxide. Samples of rock phosphate with a surface area of 7.29 m2/g were used in the sorption experiments, where the sorbent dosage was 5 mg/mL. Ternary solutions of Pb+2-Cu+2-Zn+2 were employed to obtain kinetic and equilibrium sorption data at 30 °C. Results indicated that the removal of each ion was suppressed by the presence of other competing metal ions. In fact, the sorption capacity of low-grade phosphate rock decreased in multimetallic system in comparison with a single component solution. Kinetic and equilibrium data were correlated using classical monocomponent pseudo-first and second order models, while Langmuir and Freundlich models were used for isotherm fitting. It appears that ion exchange coupled with complexation mechanism is the most probable mechanism responsible for metal uptake by low-grade rock phosphate. On the other hand, Boujelben et al. [33] studied the feasibility of using natural iron oxidecoated sediment as sorbent for the removal of Cu+2 and Ni+2. The selection of this material was based on its low cost and abundance in Tunisian ores. The surface area of this solid was 6.97 m2/g and it was characterized by some techniques as FTIR spectroscopy, energy dispersive of X-ray and X-ray diffraction. The results of elemental analysis indicated the presence of silica, iron oxides and other minerals. Binary sorption experiments were performed at pH 5 and 20 °C using different metal concentrations (0.5 – 1.9 mmol/L) and a sorbent dosage of 20 mg/mL. The sorbed amount of metal ions in the mixed solutions was less than those for single-component solutions showing an antagonistic competitive effect between these metallic species. In fact, the removal of Cu+2 was greater than Ni+2, suggesting that the functional groups on the surface of this sediment had a relatively stronger affinity for Cu+2 than Ni+2. This study concluded that the natural iron oxide-coated sediment may be used for the simultaneous removal of Cu+2 and Ni+2 ions from metal-containing effluents.

2.8. Lignin The use of Kraft lignin as an alternative low-cost sorbent for heavy metal removal has been considered to solve the disposal problem associated to this industrial waste [12]. Lignin

Water Production and Wastewater Treatment, Nova Science Publishers, Incorporated, 2010. ProQuest Ebook Central,

Multicomponent Removal of Heavy Metals…

79

is recognized as the second most abundant natural polymer after cellulose and comprises from 17 to 30% of plant biomass. In particular, lignin isolated from wastes of Kraft pulping has a high surface area (≅ 1260 m2/g) and presents polyhydric phenols and sulfur-containing groups that may act for the removal of heavy metals. Mohan et al. [12] have reported the removal of Cu+2, Cd+2 and Zn+2 in binary and ternary metal solutions using Kraft lignin at pH 4.5 and 25 °C. It is convenient to mention that authors only reported and analyzed the sorption behavior for Cd+2 and Cu+2 ions, and the sorption capacities for Zn+2 were not provided. Besides, the sorption tests were performed at different metal concentrations (0.01 – 5 mmol/L) using a Cu+2 to Cd+2 ratio of 1:1. Results of this study shown that metal sorption capacity for both Cd+2 and Cu+2 decreased more in ternary mixtures in contrast with the binary systems. For all studied systems, there was sorption suppression by the presence of other metallic species. The sorption capacity of lignin for Cu+2 ions in the binary and ternary systems was always significantly greater than the other metal ions. It appears that the presence of Zn+2 affects the removal of both Cd+2 and Cu+2 ions, being more significant the competitive effect for Cd+2. Besides, in the binary system Cd+2-Cu+2, lignin preferred Cu+2 to Cd+2 ions. These relative sorption preferences were related to the relative ionic property orders (e.g. hydrated ionic radius, electronegativity and standard reduction potential). The correlation of multicomponent sorption data was performed using the classical (i.e., monocomponent) Langmuir and Freundlich models where the Langmuir isotherm provided the best fit. Authors suggested that several mechanisms are involved in the multicomponent heavy metal removal using lignin, which may include sorption, surface deposition and ion exchange.

Copyright © 2010. Nova Science Publishers, Incorporated. All rights reserved.

2.9. Olive Wastes The production of olive oil generates large volumes of wastes that vary in composition depending on the production system [34-36]. Current studies have proposed the waste of olive oil production as sorbent of heavy metals. For example, Pagnanelli et al. [34] studied the simultaneous removal of Cu+2 and Cd+2 ions using olive pomace, which is a solid waste from olive oil production. These authors noted that the oil production yields a solid residue (olive pomace) (30%), which may be used as fertilizer, nutritive additive for animal food, source of heat energy, soil stabilizer, and also as low-cost sorbent. Olive pomace consists of cellulose, lignin and uronic acids along with oily wastes and polyphenolic compounds. This complex matrix contains numerous functional groups useful for metal binding. The behavior of this sorbent was tested using binary solutions of Cu+2 and Cd+2 ions with different metal concentrations (0.3 – 1.8 mmol/L) at different pH levels. Overall, the sorption of each metal decreased as the initial concentration of the antagonist metal in solution increased. This effect was more evident for Cd+2 sorption while little effect can be noticed for Cu+2 at tested conditions. The competition effect was more evident at pH 5 than 4 for both Cd+2 and Cu+2 ions. A statistical analysis indicated that Cu+2 ions significantly influence Cd+2 removal at both pH levels. Olive pomace may bind metallic species by different mechanism such as complexation, chelation, physical sorption, ion exchange and electrostatic interactions. On the other hand, Martinez-Garcia et al. [35] reported the application of olive mill waste for the sorption of heavy metals from aqueous solution. This waste has a high moisture content (55 – 60%) and contains polyphenolic compounds, which limit disposal alternatives. Multicomponent sorption experiments were performed using binary and ternary metallic

Water Production and Wastewater Treatment, Nova Science Publishers, Incorporated, 2010. ProQuest Ebook Central,

80

H. E. Reynel-Avila, D. I. Mendoza-Castillo, V. Hernández-Montoya et al.

solutions of Pb+2, Cd+2 and Cu+2 ions at 20 °C and different conditions of pH. This study noted that when Cu+2 was in solution with other metal ions, its uptake was considerably reduced. Also, Pb+2 sorption presented the same adverse effect when it was mixed with other metallic species. Authors concluded that the uptake ability of olive mill waste was seriously compromised resulting in low sorption capacities for tested heavy metals. Finally, another study reported the reuse of olive stone waste in its native form without further treatment as sorbent material for the removal of Cu+2, Cd+2, Ni+2 and Pb+2 from binary mixtures [36]. An olive stone waste with a specific surface area of 0.187 m2/g and obtained from pulp generated in the oil production industry was used in the sorption studies. Batch experiments were carried out at 20 °C, initial pH of 5.5 and using binary metal mixtures in equimolar concentrations. Authors noted that metals in binary mixtures showed an antagonistic competitive effect; however, multicomponent sorption isotherms were not reported with exception of Cd+2-Pb+2 system. They suggested that sorption-complexation in addition to ion-exchange may be involved in the case of sorption of the Cu+2, Cd+2 and Pb+2 ions, while ion-exchange appears to be the most important mechanism for Ni+2 sorption.

Copyright © 2010. Nova Science Publishers, Incorporated. All rights reserved.

2.10. Orange Wastes Pérez-Marín et al. [37] reported the sorption of Cd+2, Zn+2 and Pb+2 ions in binary mixtures using orange wastes as sorbent. Orange waste consists of cellulose, hemicellulose, pectin, and limonene. FTIR analysis indicated the presence of metal binding groups such as carboxyl and hydroxyl. This waste was collected from an orange juice manufacturer, washed with tap water and finally dried. Binary sorption experiments were performed at pH 4 using the subsequent addition method with initial metal concentrations from 0.07 to 1.5 mmol/L. For each binary system studied (Cd+2-Zn+2, Cd+2-Pb+2, Pb+2-Zn+2), an antagonistic sorption behavior was determined where the sorbed amount of one ion decreased when the concentration of the co-ion increased. Specifically, the removal of Cd+2 and Zn+2 was moderately affected by the presence of the other metal. For binary mixtures with Pb+2, the uptake of Cd+2 and Zn+2 decreased significantly, especially at high Pb+2 concentrations. However, the presence of Cd+2 and Zn+2 ions in solution has a small effect on the capacity of orange waste to remove Pb+2 ions. The affinity of orange wastes for these heavy metals was related to the physicochemical parameters of metallic species: atomic weight, ionic radius, possible coordination number, and electronegativity. Finally, the extended Langmuir model was successfully used for data regression.

2.11. Peat Peat has been recognized as a low-cost sorbent for removal of heavy metal ions from wastewaters [38]. It is a naturally abundant and inexpensive material, which is constituted by lignin, cellulose, humic and fulvic acids. Studies reported in literature indicate that peat contains polar functional groups such as aldehydes, ketones, carboxylic, hydroxylic and phenolic groups, that can be involved in the heavy metal uptake via sorption. Besides, peat is also known by its ion exchange properties [13,39,40]. In particular, the presence of humic acid in the peat is considered as the responsible of the metal ion sorption. Probably, peat is the

Water Production and Wastewater Treatment, Nova Science Publishers, Incorporated, 2010. ProQuest Ebook Central,

Copyright © 2010. Nova Science Publishers, Incorporated. All rights reserved.

Multicomponent Removal of Heavy Metals…

81

most studied low-cost sorbent for removal of heavy metals in multicomponent systems and the sorption behavior of peat has been tested especially in binary and ternary systems [13,3843]. For example, McKay and Porter [39] reported the results for multicomponent equilibrium studies of Cd+2-Cu+2, Cd+2-Zn+2, Cu+2-Zn+2 and Cd+2-Cu+2-Zn+2 mixtures on peat. Sorption experiments were determined at 20 °C and pH 4.5, using metal solutions with different initial concentrations. By comparing the single and multimetallic sorption capacities, competition effects can be observed between the different metal species in solution. In fact, the sorption capacity of Cu+2 in the binary and ternary systems was always significantly greater than the other metal ions. This study showed that Cu+2 has a significant impact on sorption behavior of both Zn+2 and Cd+2. This study suggested that the sorption mechanism for Cu+2 is different from that of Cd+2, Zn+2 and other divalent metals. On the other hand, Ho et al. [41] and Ho and McKay [42] studied the competitive sorption of Cu+2 and Ni+2 onto peat. Specifically, Ho et al. [41] reported the effect that a competing ion has on the rates of removal of Cu+2 and Ni+2 ions at pH 5 and 25 °C, while Ho and McKay [42] studied the effects of competitive sorption in batch systems at various metal ion concentrations of Cu+2-Ni+2 at pH 4.5 and 25 °C. In bi-solute kinetic studies, the presence of secondary metal affects the sorption of both Cu+2 and Ni+2, and the equilibrium concentrations of both metals were lower than in their respective mono-solute systems. It appears that Cu+2 has a greater effect on the Ni+2 uptake in the bi-solute systems causing a decrease in the diffusion rate. Ho et al. [41] established that a pseudo-second order model provides a more appropriate description of the bi-solute binding of Cu+2 and Ni+2 ions to peat. The results of equilibrium sorption studies of Ho and McKay [42] also indicated that competition occurred in Cu+2-Ni+2 system and Cu+2 is removed more extensively than Ni+2. The predicted equilibrium data using Langmuir-based models were found to be in agreement with experimental values. Peat has also been used for simultaneous removal of Cr+3, Cd+2 and Cu+2 ions in binary metal solutions at pH 4 and 22 – 25 °C [38]. As expected, the presence of co-ions decreased the metal uptake and this competitive effect increased as co-ion content was increased. This study indicated that the percentage reduction in metal uptakes ranged from 50 to 70% at the highest co-ion concentrations. However, Cd+2 as a co-ion diminished Cr+3 uptake by a maximum of 30%, whereas Cu+2 uptake was reduced by approximately 50% in the presence of the same Cd+2 concentration. Both Cr+3 and Cu+2 caused large and similar reductions in Cd+2 uptakes. This study concluded that the similarity of the uptake competition effects of Cr+3 and Cu+2 is noteworthy and is likely to result from the likeness of their ionic radii. Qin et al. [13] and Qin and Wen [44] studied the competitive sorption of Pb+2, Cu+2 and Cd+2 in binary and ternary systems. Both studies concluded that Pb+2 was always favorably sorbed on peat over Cu+2 and Cd+2, and Cu+2 over Cd+2 in the multisolute systems. The individual sorption capacities of the three metallic species decreased in binary and ternary systems, where the decrease in sorption capacity was greater in ternary solute system as compared to binary systems. In fact, Cd+2 sorption decrease was more significant in binary and ternary systems than Pb+2 and Cu+2. Pb+2 ions significantly inhibited the removal of Cu+2 and Cd+2, and Cd+2 was more affected in the competitive sorption. Hence, the sorption capacity of Pb+2 was reduced to a less extent than that of Cu+2 and Cd+2. These studies indicated that as the metal concentrations increased further, metals with higher affinity competed with metal with lower affinity for the sorption sites of peat. When metal competes

Water Production and Wastewater Treatment, Nova Science Publishers, Incorporated, 2010. ProQuest Ebook Central,

82

H. E. Reynel-Avila, D. I. Mendoza-Castillo, V. Hernández-Montoya et al.

for the same sorption sites of a sorbent, metal with a greater affinity could displace others with a lower affinity. Multicomponent kinetic studies showed typical biphasic kinetics with a rapid sorption of Pb+2, Cu+2 and Cd+2 on peat at the initial stage followed by a slower one. The second-order model fitted the kinetics of these ions in multicomponent systems obtaining satisfactory correlation coefficients. Finally, Qin et al. [13] reported the application of X-ray absorption spectroscopy for studying the mechanisms of competitive removal of heavy metals on peat. Their results showed that both Cu+2 and Pb+2 were bonded primarily to carboxylic ligands of peat without excluding hydroxylic groups, providing an evidence for the competitive behavior of these metallic species. Recently, Balasubramanian et al. [40] reported the multicomponent sorption of Pb+2, Zn+2 and Cd+2 ions onto Indonesian peat at 23 °C and pH 6. From results obtained using binary and ternary metallic solutions, it was observed that Pb+2 was favorably sorbed onto peat over Cd+2 and Zn+2, and Cd+2 over Zn+2. The sorption capacities for three metallic species showed decreases in both binary and ternary systems, and the decrease was greater in the ternary systems as compared to the binary system. This decrement in sorption capacity was associated to the metal interactions and sorption antagonistic behavior. These authors concluded that in addition to ion exchange, a surface sorption–complexation mechanism may be involved in heavy metal removal using peat. In general, carbon-oxygen surface function groups were considered to be responsible for the sorption of heavy metal ions. It is convenient to note that the possible differences in competitive effects and sorption preferences may be related to the great variety of peat types.

2.12. Pine Bark Pine bark, another agricultural waste, has been used for multicomponent removal of Cd+2, Ni , Cu+2 and Pb+2 ions in binary, ternary and quaternary metallic solutions [45]. Sorption capacities of bark and other agricultural materials are generally attributed to their protein, carbohydrates and phenolic compounds, which have metal-binding functional groups such as carboxyl and amino groups. Pine bark from Ottawa region was used for the sorption tests without treatment and the experiments were performed at pH 4, in which each metal had an initial concentration of 100 mg/L. The interaction effects in multicomponent sorption experiments using bark were identified and the authors reported that in most cases the sorption capacity for one ion was significantly affected by the presence of the others. It appears that Pb+2 ions suppressed the sorption of both Cu+2 and Cd+2 ions in ternary and binary systems. On the other hand, Ni+2 sorption was strongly inhibited by Cu+2 ions in the ternary and quaternary systems that contained these metals. However, Ni+2 inhibited the removal of both Cu+2 and Pb+2 ions in binary solutions. It is interesting to note that synergism effect was reported for Ni+2 ions in several binary mixtures at tested conditions. As expected, the competitive effect in quaternary mixture is higher than those reported for ternary and binary mixtures. These results suggest a complex interaction of heavy metal ions on pine bark under competitive conditions. Overall, this study concluded that the following affinity series may be established using binary selectivity indexes: Pb+2 > Cu+2 > Cd+2 > Ni+2. The principal mechanisms for heavy metal removal using pine bark may be due to the sorption on surface and pores, and complexation [45].

Copyright © 2010. Nova Science Publishers, Incorporated. All rights reserved.

+2

Water Production and Wastewater Treatment, Nova Science Publishers, Incorporated, 2010. ProQuest Ebook Central,

Multicomponent Removal of Heavy Metals…

83

2.13. Sewage Sludge

Copyright © 2010. Nova Science Publishers, Incorporated. All rights reserved.

Sewage sludge is a natural carbonaceous material with volatile components in their structure. By this reason, this material has been considered as potential precursor for the production of sorbents. For example, Otero et al. [46] have reported the study of two different sewage sludge-based sorbents for the removal of heavy metals under competitive conditions. Sewage sludge from an urban treatment plant using activated sludge biological treatment was used for the production of sorbents by pyrolysis. This sludge was pyrolized with and without previous chemical activation. These carbonized sewage sludges were used for multicomponent sorption experiments. Authors indicated that surface area of the chemically activated carbon (472 m2/g) was higher than the carbon prepared without activation (60 m2/g). Sorption experiments were performed in binary solutions of Hg+2-Pb+2, Hg+2-Cu+2, Hg+2-Cr+3, Pb+2-Cu+2, Pb+2-Cr+3, and Cu+2-Cr+3 at 25 °C and pH 3 – 5. Results indicated that the carbonized sewage sludge with chemical activation has a higher capacity to remove these metal ions. In all cases, the presence of a competitive metal in solution caused a reduction of the sorption capacity compared with the obtained in single sorption. The presence of Cr+3 affects less the sorption of Pb+2 and Hg+2 than any other metal ion. The sorption of Pb+2 was remarkably reduced by Hg+2. The Cr+3 is the less competitive metal for Pb+2 removal. Note that Cu+2 sorption was strongly inhibited by Hg+2 ions while the presence of Pb+2 ions in solution had a small effect on Cr+3 and Cu+2 sorption. Conventional Langmuir isotherm was used for data modeling in these binary metallic solutions. The sorption capacity in competition onto carbonized sewage sludge was Hg+2 > Pb+2 > Cu+2 > Cr+3.

3. COMPARISON OF MULTICOMPONENT SORPTION CAPACITIES OF LOW-COST SORBENTS UNDER COMPETITIVE CONDITIONS AND THE SORBENT REGENERATION 3.1. Sorption Capacities The review indicates that a wide variety of sorbents has been studied for removal of metallic species in multicomponent systems. Many of these materials have not cost, are abundant, easily available and may be biodegradable. Their use as sorbents for water treatment provides an aggregated value. They are usually used without any treatment but it is feasible to modify their morphology, texture, surface groups, and chemical composition for enhancing the sorption capacity. It is convenient to remark that the improvement of sorption capacity in these sorbents using a proper treatment may not imply a significant increase of its final cost. With illustrative purposes, Figures 1 – 3 and Table 3 show the highest sorption capacities of metals in aqueous solutions (qe) for several sorbents. These capacities have been standardized and are reported in molar basis for a direct comparison because the analysis of sorption behavior may be misinterpreted when they are on weight basis [12]. Literature may fail to provide the detailed experimental conditions where the sorbent behavior has been tested, and these conditions may vary significantly from one study to another. Therefore,

Water Production and Wastewater Treatment, Nova Science Publishers, Incorporated, 2010. ProQuest Ebook Central,

Copyright © 2010. Nova Science Publishers, Incorporated. All rights reserved.

84

H. E. Reynel-Avila, D. I. Mendoza-Castillo, V. Hernández-Montoya et al.

these sorption capacities should be considered as a reference of values that can be achieved under a specific set of conditions rather than a maximum sorption behavior. As expected, the competitive effect in sorption process is more noticeable with the increase in the number of metal ions in solution and the sorbent behavior may vary significantly for different co-ions (see results reported in Figures 1 – 3 and Table 3). Overall, antagonistic competitive effects occur for various sorbents and different metallic systems. Note that the majority of sorption data is focused on equilibrium studies in binary solutions. In fact, Tables 2 and 3 indicate that there is a lack of sorption studies employing solutions with three or more co-ions. Binary solutions with Cd+2 and Cu+2 as co-ion have been widely used to assess competitive effects on the sorption behavior of several materials. The sorption capacities of Cd+2 and Cu+2 metal ions in aqueous solution using some low-cost sorbents are given in Figures 1 and 3, respectively. The industrial use and pollutant potential of these metals is perhaps an explanation of their selection as benchmark in sorption studies. It appears that the presence of Cu+2 in multicomponent solutions has a significant effect on sorption of other co-ions. Although there are significant differences in the sorbent behavior, chitosan and alginate have demonstrated outstanding sorption capacities in multimetallic systems. Peat may also offer an attractive behavior for some binary solutions (see Figures 1 – 3). Note that several authors have attempted to use various physical and chemical properties of metal ions such as ionic radius, electronegativity, electron affinity, chemical coordination characteristics, and solvatation to explain and justify the affinity and sorption preferences in multimetallic solutions [19]. Although the trends of multimetallic sorption behavior may show qualitatively dependence on certain of these chemical properties, at present it is difficult to identify common factors because several variables affect the sorbent behavior. Specifically, the sorption preference and affinity for different heavy metals are related to the characteristics of the sorbent (e.g., functional groups, structure, textural properties, etc), the solution chemistry (e.g., pH, temperature, ionic strength), and the presence of diverse sorption mechanisms. Undoubtedly, the identification of factors affecting the selectivity and affinity of low-cost sorbents is a future research topic. In this context, it is convenient to remark that the multicomponent isotherms are useful to characterize the sorbent capabilities when several metallic species are present. They are also important to explore systematically the effect of the concentration of one metal ion on the uptake of other metal ions and to identify the effect of pH, temperature, and ionic strength on sorbent behavior. For example, the application of three-dimensional sorption surfaces for binary systems allows the understanding of competitive effects of solutes, and the extrapolation and interpolation of metal uptakes [38].

Water Production and Wastewater Treatment, Nova Science Publishers, Incorporated, 2010. ProQuest Ebook Central,

Water Production and Wastewater Treatment, Nova Science Publishers, Incorporated, 2010. ProQuest Ebook Central,

Copyright © 2010. Nova Science Publishers, Incorporated. All rights reserved.

Table 3. Sorption capacities and competitive effects reported for several low-cost sorbents in aqueous solutions of ternary, quaternary and quinary metal systems Sorbent Ternary mixture Bone char Chicken feathers Crab shell Lignin Low grade phosphate rock Peat

Pine bark

Sorption capacity, mmol/g 1 Cd+2 Cu+2 Cr+3 Zn+2 0.200 0.100 0.143

0.040 0.050 0.116 0.023 0.047 0.028

0.430 0.007

1

Ni+2

Hg+2

0.150 0.003 0.470

0.500 0.152

NS

0.148

0.076

0.080 0.320

0.030 0.138

0.066 0.099

0.089 Rice husk ash 0.023 Volcanic lava ash 0.007 0.009 Quaternary and quinary mixture Pine bark 0.038 0.061 Chicken feathers 0.002 0.008

Pb+2

0.003

0.045 0.450 0.106 0.041 0.042 0.039

0.070

0.002 0.058 0.0 0.062

0.005

0.001

0.040 0.004

NS indicates that sorption capacity or competitive effect is not specified.

0.0 0.002

Antagonism

Synergism

√ NS √ (Cd+2) √ (Cu+2,Cd+2)

Noninteraction

√

Ref. 23 25 30 12

√

32

√ √ √ √ √ √ √ √ NS

13,39

√ NS

45 +2

√ (Cu ) √ (Ni+2) 10 19 45 25

86

H. E. Reynel-Avila, D. I. Mendoza-Castillo, V. Hernández-Montoya et al. A: Antagonism

1.6

S: Synergism

+2

N: Non-interaction +2

+2

Cd +2 Ni

Cd +2 Cu

Cd +2 Zn

qe, mmol/g

1.2 A A

0.8

A

A

0.4 A

A

A

N S

A

A

A A

A

A

A

A

A

A A A

A A

A

S

Ba ga ss ef ly as Ri h ce hu sk as h Pi ne Ri ba ce rk hu sk Bo as ne h ch ar O ra Lig ng n e w in as te s Pe a t Al g Bo ina n e te ch O liv Li ar g e p ni om n ac e P Pi e ne a t ba rk

0

AA

1.6

+2

N

+2

Cd +2 Pb

Cd +3 Cr

qe, mmol/g

1.2 0.8 A

A AA

A

0.4

N A

AA A

A A A

Figure 1. Sorption capacities and competitive effects reported for several low-cost sorbents in aqueous solutions of binary metal systems with Cd+2 as principal co-ion. A: Antagonism

0.8

qe, mmol/g

S: Synergism

N: Non-interaction +2

+2

Pb +2 Zn

Pb +3 Cr

0.6 A

A

0.4 A

0.2

A A A

A A

Pe at

wa ste s Or an ge

slu dg e Se wa ge

sh ell

0

Cr ab

Copyright © 2010. Nova Science Publishers, Incorporated. All rights reserved.

O

Pe at

Al Cr gina ra ab S te ng h e w ell as te s P P i ea ne t ba rk Cr ab sh el l

0

Figure 2. Sorption capacities and competitive effects reported for several low-cost sorbents in aqueous solutions of binary metal systems with Pb+2 as principal co-ion.

Water Production and Wastewater Treatment, Nova Science Publishers, Incorporated, 2010. ProQuest Ebook Central,

Multicomponent Removal of Heavy Metals…

A: Antagonism

2.8

S: Synergism +2

87

N: Non-interaction +2

+2

Cu +3 Cr

Cu +2 Zn

Cu +2 Pb

qe, mmol/g

2.1 A

A

1.4 A

AA

A

A

A A

qe, mmol/g

A

A

AA

A A

+2

+2

Cu +2 Ni

Cu +2 Hg

S A

2.1

A

Al gi na te Pe at Pi Se ne b wa a ge rk slu Ch Bo dge ick ne en ch fe ar ath Ch ers ito sa Li n gn in Pe at

Se wa ge

slu dg e

Pe at 2.8

A

A

A

0.0

S

1.4 0.7

A

A

A

A A

A A

A A

A

A

A S

Gl ut ar C Ep ald ich ehy hito sa d lo n ro e/c hi hy t dr os in an /c Se hito w ag san es C lu Iro hic dg ke e n n ox fe id a eco ther s ate d sa nd Pe at Pi ne ba rk

0 Copyright © 2010. Nova Science Publishers, Incorporated. All rights reserved.

A

A

0.7

Figure 3. Sorption capacities and competitive effects reported for several low-cost sorbents in aqueous solutions of binary metal systems with Cu+2 as principal co-ion.

However, the obtaining of multicomponent isotherms at different operating conditions may require a long and tedious experimental procedure especially for multisolute systems. As discussed in previous section, some studies may fail to obtain the multicomponent isotherms and, as a consequence, the competitive effects of all metal ions can not be clearly identified. In these conditions, experimental designs can be used to assess the sorbent behavior in systems with various metallic species. For instance, Pérez-Marin et al. [37] introduced alternative experimental strategies based on an aggregation method and an experimental design for easily obtaining sorption data in multicomponent systems. This kind of experimental approach is very useful for characterizing the sorbent behavior under competitive conditions.

Water Production and Wastewater Treatment, Nova Science Publishers, Incorporated, 2010. ProQuest Ebook Central,

88

H. E. Reynel-Avila, D. I. Mendoza-Castillo, V. Hernández-Montoya et al.

3.2. Regeneration of Low-Cost Sorbents The sorbent regeneration is also an important issue to establish the feasibility of new materials in wastewaters treatment. Several desorbing agents have been tested for metal recovery and sorbent regeneration and they include both inorganic and organic agents such as HCl, HNO3, H2SO4, CH3COOH, EDTA, and NaOH. The behavior of these desorbing agents depends on the sorbent type, the concentration of sorbed metals, and the operating conditions (e.g., concentration of desorbing agent, temperature, and sorbent dosage). Overall, the literature indicates that aqueous solutions of HCl and HNO3 are effective for metal desorption in several low-cost sorbents. However, some studies have suggested that the complete regeneration of sorbent used for heavy metal removal is not possible due to the incorporation of metal into the sorbent structure, re-sorption process and other factors [14,17].

Copyright © 2010. Nova Science Publishers, Incorporated. All rights reserved.

4. MODELING OF MULTICOMPONENT SORPTION OF HEAVY METALS As mentioned in previous sections, the sorption data generally implies experimental information obtained from equilibrium and kinetic studies. Kinetic experiments are performed to establish the equilibrium time, to study the rate of solute uptake, and to establish the ratecontrolling step; while equilibrium sorption studies are necessary to provide the maximum sorption capacity of the sorbent, to calculate physico-chemical parameters of sorption process, and to determine competitive effects in multicomponent systems. The equilibrium relationships between sorbent and sorbate are described by sorption isotherms, which are usually represented by the relationship between the quantity sorbed and the sorbate concentration remaining in the solution at a fixed operating conditions (i.e., pH, temperature, ionic strength, and others). In order to predict the sorbent behavior under competitive conditions, it is necessary to carry out modeling studies based on multicomponent equilibrium and kinetics studies. Thus, data collection in the study of sorption process is normally followed by regression of the experimental information using theoretical or empirical models with the aim of developing mathematical equations that describe satisfactorily the system under analysis. These models can be used for interpolation or extrapolation of sorption behavior, to investigate competitive mechanisms, and to calculate parameters useful for the design, optimization and control of water treatment processes. The modeling of multicomponent equilibrium sorption has been a research topic during many years. Actually, several models have been proposed and applied to correlate multicomponent sorption data for heavy metal removal. In the literature, the modeling of multisolute sorption can be approached by using classical multicomponent equations, usually considered as empirical models, or by developing chemico-physical mechanistic models able not only to represent but also to explain and predict the experimental sorbent behavior [34,47]. Note that a major understanding of the chemical and physical aspects involved in multicomponent sorption process is possible with the use of mechanistic models instead of empirical ones. Mechanistic models suppose the presence of different types of metal binding sites on the sorbent surface and consider specific reactions between these sites and the metal

Water Production and Wastewater Treatment, Nova Science Publishers, Incorporated, 2010. ProQuest Ebook Central,

Multicomponent Removal of Heavy Metals…

89

species in solution, where different reaction mechanisms can be used. These models require, at least, a preliminary sorbent characterization and knowledge of the particular solution chemistry of the solutes under study [34,47]. Unfortunately, due to the complexity of sorption behavior and the potential presence of several mechanisms for heavy metal removal using low-cost sorbents, complex mechanistic models are required for a successful data modeling. In addition, the development of these models may involve wide and deep experimental studies and independent sets of data to give real consistency to the model parameters and also to avoid that a better fitting was reached only by the introduction of additional adjustable parameters [47]. By this reason, it is a common practice to use empirical models for multimetallic sorption modeling due to their flexibility and simplicity both in formulation and computer implementation, which are attractive characteristics for their application in the design of wastewaters treatment processes. In the following sections, we briefly describe the classical multicomponent models used for competitive sorption data analysis as well as different key aspects for performing a reliable data regression.

Copyright © 2010. Nova Science Publishers, Incorporated. All rights reserved.

4.1.

Description of Classical Multicomponent Sorption Models

Traditional sorption equations employed for representing single solute systems have been used to develop several multicomponent sorption models. They are simple mathematical expressions with relatively few adjustable parameters, generally two or three parameters for representing the sorption behavior of each solute. The model parameters can be fitted both on single and multicomponent experimental data with sufficient accuracy for describing the complexity of competitive sorption data for heavy metal removal. In literature, the models for multicomponent sorption are generally considered as predictive and competitive models [34,47]. In predictive models, it is frequently assumed a hypothetical sorption behavior where the sorption behavior of solutes does not depend on other species in solution and, as a consequence, the model parameters are regressed by using only single component data. On the other hand, the concept of non-ideal competition among several solutes in solution assumes that each species may modify its sorption behavior if other competitive species are present or not. Therefore, the model parameters for competitive models are usually obtained from data fitting of multicomponent sorption experiments or by combination of both single and multicomponent data. Note that multicomponent sorption models can be made more flexible by inserting additional correction parameters that take into account the interaction between the solutes in solution and these parameters are fitted on multicomponent sorption data. These correction factors usually improve the results of data regression and are incorporated in the model by correcting the individual sorbate concentration using an interaction term, which is assumed as a characteristic of each species and depends on the concentration of the other components in the solution [14]. Table 4 describes a set of representative models used for the correlation of sorption isotherms in multicomponent systems. Several multicomponent isotherm models have been derived from the classical isotherms equations by keeping the same assumptions made in the treatment of pure component in the sorption equilibrium. These models include Langmuirbased, Sips, Freundlich, and Redlich-Peterson multicomponent equations. For example, extended, non-modified and modified Langmuir models are extensions of the basic Langmuir

Water Production and Wastewater Treatment, Nova Science Publishers, Incorporated, 2010. ProQuest Ebook Central,

90

H. E. Reynel-Avila, D. I. Mendoza-Castillo, V. Hernández-Montoya et al.

isotherm [17,18]. Characteristics and details of some models can be found in studies reported by Srivastava et al. [14,17,18], Pagnanelli et al. [34] and Pagnanelli et al. [47]. It is convenient to remark that the names of multicomponent sorption models are a matter of debate. Therefore, in the context of this chapter, the nomenclature adopted in Table 4 is used for discussion of the model behavior in multicomponent sorption studies. This nomenclature is consistent with the reported by Srivastava et al. [14,17,18] and Ahmadpour et al. [48]. In general, the derivatives of Langmuir and Freundlich models have been widely used in several studies because they can be easily incorporated in a complete process model and may offer a good fit using few adjustable parameters (see results reported in Table 5). Specifically, Langmuir and Freundlich-based models have been applied for data fitting of multicomponent sorption using bagasse fly ash, rice husk ash, chicken feathers and other sorbents. However, it is convenient to remark that there is no general model applicable to all sorbate/sorbent systems due to the wide range of systems leading to a wide variety of equilibrium behavior. On the other hand, few attempts have been performed for modeling multicomponent kinetic sorption data in comparison with those studies related to equilibrium behavior. Typically, multicomponent kinetics studies are performed to determine the equilibrium time and the modeling, if any, is carried out using a predictive approach where the classical kinetic equations and their parameters used for single component systems are applied in multicomponent systems, e.g. Prasad et al. [32] and Ho et al. [41].

Copyright © 2010. Nova Science Publishers, Incorporated. All rights reserved.

4.2.

Determination of Adjustable Parameters of Multicomponent Sorption Models

In general, the adjustable parameters of multicomponent sorption models can be obtained using the following approaches: a) all parameters are determined from single sorption experiments and the data fitting in multisolute systems is not required, b) some parameters are derived from mono-component data but also there are adjustable parameters that take into account the interaction between the solutes in solution and that are fitted on multicomponent sorption data, and c) all model parameters are fitted using sets of multicomponent experimental data. The first approach is the most used for modeling the heavy metal removal using low-cost sorbents and employs multicomponent models containing only parameter values obtained from single metal experiments. This approach avoids any fitting on multicomponent experimental data. However, several studies indicated that this approach may fail to describe the competitive effects between metal ions and usually provides poor estimations for multicomponent sorption equilibria [16].

Water Production and Wastewater Treatment, Nova Science Publishers, Incorporated, 2010. ProQuest Ebook Central,

Water Production and Wastewater Treatment, Nova Science Publishers, Incorporated, 2010. ProQuest Ebook Central,

Copyright © 2010. Nova Science Publishers, Incorporated. All rights reserved.

Table 4. Multicomponent isotherm models used for the sorption data regression with low-cost sorbents Isotherm name Extended Langmuir

Model

q e ,i =

q max K i C e ,i c

1 + ∑ K j C e, j j =1

Non-modified Langmuir

Modified Langmuir

q e ,i =

q m ,i K L , i C e ,i c

1 + ∑ K L, j C e, j j =1

qe ,i =

qm ,i K L ,i (Ce ,i / nL ,i ) c

1 + ∑ K L , j (Ce , j / nL , j )

Description

Ref.

This model assumes that the surface sites are uniform and that all the sorbate molecules in solution are competing for the same surface sites (i.e., a unique qmax for the different species in competition). The model parameters are qmax and Ki.

14

This model is an extension of the basic Langmuir isotherm and is characterized by specific adjustable parameters (qm,i and KL,i) for each species in competition.

14

In this model, the interactive effect between different sorbates is incorporated by correcting the individual sorbate concentration by an interaction term nL,i.

14

This model is also an extension of the conventional Sips isotherm and is characterized by specific adjustable parameters (as,i, bs,i and ns,i) for each species in competition. This equation corresponds to a special case of surface energetic heterogeneity.

48

This model is obtained by applying the ideal adsorption solution theory (IAST). The adjustable parameters are Ks,i, bi and ni for each species in solution.

48

This model assumes that each component individually obeys the Freundlich isotherm where an exponential distribution of site sorption energies exists. The adjustable parameters are aij, KF,i and nFi where competition coefficients aij account for the inhibition to the sorption of the component i by the component j.

54

This model is an extension of the Redlich-Peterson isotherm and is characterized by specific adjustable parameters (KR,i, aR,i and βi) for each species in competition.

17

This model is a modification of multicomponent Redlich-Peterson isotherm, which is obtained by introducing an interaction term nR,i.

17

j =1

1 / n s ,i

Non-modified Sips

qe , i =

a s , i Ce , i c

1 + ∑bs , j C

1 / ns , j e, j

j =1

1 / ni −1

IAST-Sips

⎛ c ⎞ K s ,i bi C e ,i ⎜⎜ ∑ b j C e , j ⎟⎟ ⎝ j =1 ⎠ q e ,i = 1 / ni ⎛ c ⎞ 1 + ⎜⎜ ∑ b j C e , j ⎟⎟ ⎝ j =1 ⎠

Sheindorf-RebuhnSheintuch (Freundlich-type)

⎛ c ⎞ qe,i = K F ,i Ce,i ⎜⎜ ∑ aij Ce, j ⎟⎟ = 1 j ⎝ ⎠

Non-modified Redlich-Peterson

qe,i =

nFi −1

K R ,i Ce,i c

1 + ∑ aR , j Ce, jj β

j =1

Modified RedlichPeterson

qe ,i =

K R ,i (Ce ,i nR ,i ) c

1 + ∑ a R , j (Ce , j nR , j ) j =1

βj

Water Production and Wastewater Treatment, Nova Science Publishers, Incorporated, 2010. ProQuest Ebook Central,

Copyright © 2010. Nova Science Publishers, Incorporated. All rights reserved.

Table 5. Summary of data modeling using multicomponent sorption equations and several low-cost sorbents Sorbent Alginate

Heavy metals Cu+2 - Cd+2 Pb+2 - Cd+2 Cu+2 - Pb+2

Bagasse fly ash

Cd+2 - Ni+2

Rice husk ash

Cd+2 - Ni+2

Cd+2 - Zn+2

Cd+2-Ni+2-Zn+2 Chicken feathers Orange wastes

Cu+2 - Zn+2 Cu+2 - Ni+2 Zn+2 - Ni+2 Cd+2 - Zn+2 Cd+2 - Pb+2 Zn+2 - Pb+2

Isotherms and trends in the best model for sorption data fit

Modeling approach

Ref.

Modified Langmuir > Non-modified Langmuir > IAST-Sips

Predictive and competitive

16

Predictive and competitive

14

Predictive and competitive

17,18

Extended Freundlich > Sheindorf-Rebuhn-Sheintuch > Extended Langmuir > Modified Langmuir > Non-modified Langmuir Extended Freudlich > Sheindorf-Rebuhn-Sheintuch > Modified Langmuir > Non-modified Langmuir > Modified Redlich-Peterson > Extended Langmuir > Non-modified Redlich-Peterson Extended Freudlich > Sheindorf-Rebuhn-Sheintuch > Modified Langmuir > Modified Redlich-Peterson > Extended Langmuir > Non-modified Langmuir > Non-modified Redlich-Peterson Sheindorf-Rebuhn-Sheintuch > Modified Langmuir > Extended Langmuir > Non-modified Langmuir > Modified Redlich-Peterson > Non-modified Redlich-Peterson

Predictive and competitive

Predictive and competitive

Freundlich-type > Non-modified Langmuir > IAST-Sips

Predictive

24

Extended Langmuir

Competitive

37

Multicomponent Removal of Heavy Metals…

93

This behavior is generally associated to the fact that the non-ideal interactions occur in multimetallic sorption process. For the case of b) and c), the data regression for determining adjustable parameters is more flexible in modeling the sorption behavior than the simple prediction by inserting parameters regressed from separate sets of data. The determination of model parameters is based on the minimization of the differences between experimental and calculated data for each sorbate. Therefore, it is necessary to establish the parameters of multicomponent models that provide the best fit to measured data using a proper objective function. Several objective functions, which are generally called error functions, can be used for sorption data fitting and the values of model parameters may be significantly affected by the choice of this objective function [49,50]. The classical objective functions used in the modeling of multimetallic sorption data are given below c ndat

(

Fobj = ∑∑ wi qijcalc − qijexp

)

2

(1)

i =1 j =1

c ndat

Fobj = ∑∑ wi qijcalc − qijexp

(2)

i =1 j =1

⎛ qijcalc − qijexp ⎞ ⎟ Fobj = ∑∑ wi ⎜ ⎟ ⎜ q exp i =1 j =1 ij ⎠ ⎝ c ndat

c ndat

qijcalc − qijexp

i =1 j =1

qijexp

Copyright © 2010. Nova Science Publishers, Incorporated. All rights reserved.

Fobj = ∑∑ wi

2

(3)

(4)

where qi,exp and qi,calc are the experimental and predicted metal uptakes for metal i, c is the number of metallic species present in solution, ndat is the overall number of experimental data used for data correlation, and wi is a weight factor. The metal uptakes for each component qi are calculated by a mass balance

qi =

(C0,i − C f ,i )V m

(5)

where C0,i and Cf,i are the initial and final concentration of metal i in the multicomponent solution, V is the solution volume used for sorption experiments, and m is the sorbent amount, respectively. The objective functions involving fractional errors (e.g., Eqs. 3 and 4) are preferred for data fitting because they weigh errors in small and large quantities equally. For all objective functions, the introduction of weight factors (wi) leads to the bias of data fitting towards a specific sorbate. It is worth noting that several studies may fail to report the objective function and numerical strategy for finding the adjustable parameters of sorption models. This aspect is crucial considering that the values of adjustable parameters depend on these aspects.

Water Production and Wastewater Treatment, Nova Science Publishers, Incorporated, 2010. ProQuest Ebook Central,

Copyright © 2010. Nova Science Publishers, Incorporated. All rights reserved.

94

H. E. Reynel-Avila, D. I. Mendoza-Castillo, V. Hernández-Montoya et al.

The sorption data modeling is usually performed by linear and non-linear regression procedures with the local optimization methods available in commercial software such as Matlab, Mathematica, Origin, Excel (MS-Solver), Statistica, SPSS, and others. Linear regression has been the classical approach to determine parameter values of sorption models [50,51]. Unfortunately, the linear regression is not useful and feasible for data fitting of multicomponent models. In this context, a non-linear regression approach should be applied to determine the parameter values for multicomponent models. Data correlation can be performed either by direct optimization of the objective function or by solving an equivalent system of non-linear equations obtained from the stationary conditions of the optimization problem. Although isotherm and kinetics models have a simple mathematical structure and relatively few adjustable parameters, the parameter estimation problems in sorption data fitting usually have non-linear and non-convex solution spaces [52]. In fact, we have to face a global optimization problem for sorption data regression, which may be difficult to solve using traditional local optimization methods (e.g., gradient-based methods or simplex optimization method) due to: a) the presence of several local minima for the objective function used as optimization criterion, b) the number of adjustable model parameters can be large especially for systems with more than three heavy metals, and c) the model parameters may vary over a wide range of the solution domain. Note that the global minimum of error function (i.e. objective function) is the best solution to the parameter estimation problem. The failure to find the globally optimal parameters for a specific model, and using locally optimal parameters instead, can have significant consequences in subsequent calculations, and may cause errors, uncertainties in equipment design and erroneous conclusions about model behavior. Moreover, fair comparisons between different models can only occur if the global optimum parameters of all models have been identified. As can be expected, finding the global minimum is more difficult that finding a local optimum and the behavior of classical optimization strategies used for non-linear regressions can be impeded by the presence of local minima [53]. In this regard, the models with higher number of parameters (e.g., models with interaction terms) are expected to be more difficult to solve than those with lower dimension due to the increase in the complexity of solution space of parameter estimation problem. Therefore, the use of reliable methods for solving parameter estimation problems in heavy metal sorption modeling is very important. Extensive research effort has been dedicated to improve the modeling of sorption isotherm and kinetics especially for single mono-component systems [49,50]. Most of the studies has reported the effect of model type and the method (i.e., linear or non-linear regression) used in deriving the model parameters. However, there is a lack of studies concerning with the application of global optimization strategies for sorption data fitting especially for multicomponent systems [52,53]. In particular, stochastic global optimization methods, also known as meta-heuristics, have shown some promise for solving parameter estimation problems [52,53]. Their features offer several advantages for solving parameter estimation problems such as generality, reliable and robust performance, little information requirement for the optimization problem to be solved, easy implementation, and reasonable computational requirements. Evidently, the global search capabilities of these solvers become more important when data modeling is performed using multiparameter models (e.g., modified Langmuir or Redlich-Peterson isotherms). Different meta-heuristic methods can be used for sorption parameter estimation and they include: simulated annealing, differential evolution, harmony search, genetic algorithms and

Water Production and Wastewater Treatment, Nova Science Publishers, Incorporated, 2010. ProQuest Ebook Central,

Multicomponent Removal of Heavy Metals…

95

particle swarm optimization. To the best of our knowledge, ISOFIT software [53] is the only free available tool for fitting sorption data in monocomponent systems that employs a stochastic global optimization method, specifically, the particle swarm optimization. Based on the authors’ experience, stochastic optimization methods may offer the best compromise between reliability and efficiency for multivariable parameter estimation problems. However, further studies are necessary to identify the relative strengths of available stochastic optimization methods for multicomponent sorption data modeling of heavy metals. Additionally, other numerical strategies such as neural networks may be used to calculate multimetallic sorption isotherms [29].

4.3. Statistical Criterions for Evaluating the Goodness of Multicomponent Sorption Data Fitting

Copyright © 2010. Nova Science Publishers, Incorporated. All rights reserved.

A variety of statistics must be used to quantify the overall quality of the data fitting and to compare the behavior of kinetic and isotherm equations used in data regression. The most common statistic measures for evaluating the overall goodness of the fit include the mean average deviation between calculated and experimental sorption capacities, the standard deviation, and the coefficient of determination R2. These statistics should be accompanied by a study of the behavior of the relative residuals ei = (qi,exp − qi,calc)/ qi,exp to identify obvious patterns (i.e., the residuals should be structureless or have a defined tendency) and to perform model comparison. Plots of residuals can be used to perform this analysis. Finally, details of alternative statistics criterions useful for data fitting of sorption models are described by Matott and Rabideau [53].

5. CONCLUSION A wide variety of low-cost sorbents has been studied for heavy metal removal from aqueous solution under competitive conditions. The use of these sorbents is recommended since they are relatively cheap or have no cost, easily available, renewable and may show high affinity for several heavy metals. Besides, the treatment of these sorbents may increase significantly their sorption behavior. Multicomponent sorption studies for heavy metal removal are important to establish the interference of one metal on the uptake of the other metal ions. In general, the literature of low-cost sorbents indicates that the competitive effects for heavy metal ions become more marked with the increase in the number of solutes in solution where the sorption of one metal ion generally interferes with that of another. However, the comparison of their sorption capacities is difficult due to the inconsistencies and discrepancies in experimental data conditions and representation. From the literature reviewed, alginate, chitosan and peat are sorbents that stand out for their sorption behavior in multimetallic systems. The modeling of multicomponent sorption process requires further investigation especially in the application of robust numerical strategies for the determination of adjustable parameters in kinetic and isotherm models. Besides, it is necessary to perform more studies for improving the sorbent regeneration and recovery of metal ions.

Water Production and Wastewater Treatment, Nova Science Publishers, Incorporated, 2010. ProQuest Ebook Central,

96

H. E. Reynel-Avila, D. I. Mendoza-Castillo, V. Hernández-Montoya et al.

REFERENCES [1] [2]

[3] [4]

[5] [6] [7] [8] [9]

Copyright © 2010. Nova Science Publishers, Incorporated. All rights reserved.

[10]

[11]

[12]

[13]

[14]

[15]

[16]

Bailey, S.E., Olin, T.J., Bricka, R.M., Adrian, D.D. A review of potentially low-cost sorbents for heavy metals. Water Research 33 (1999) 2469-2479. Wan Ngah, W.S., Hanafiah, M.A.K.M. Removal of heavy metal ions from wastewater by chemically modified plant wastes as adsorbents: A review. Bioresource Technology 99 (2008) 3935-3948. Volesky, B. Detoxification of metal-bearing effluents: biosorption for the next century. Hydrometallurgy 59 (2001) 203-216. Sud, D., Mahajan, G., Kaur, M.P. Agricultural waste material as potential adsorbent for sequestering heavy metal ions from aqueous solutions – A review. Bioresource Technology 99 (2008) 6017 – 6027. Wang, J., Chen, C. Biosorbents for heavy metals removal and their future. Biotechnology Advances 27 (2009) 195-226. Doyurum, S., Celik, A. Pb(II) and Cd(II) removal from aqueous solutions by olive cake. Journal of Hazardous Materials B138 (2006) 22-28. Demirbas, A. Heavy metal adsorption onto agro-based waste materials: A review. Journal of Hazardous Materials 157 (2008) 220-229. Babel, S., Kurniawan, T.A. Low-cost adsorbents for heavy metals uptake from contaminated water: a review. Journal of Hazardous Materials B97 (2003) 219-243. Hanzlik, J., Jehlicka, J., Sebek, O., Weishauptova, Z., Machovic, V. Multi-component adsorption of Ag(I), Cd(II) and Cu(II) by natural carbonaceous materials. Water Research 38 (2004) 2178 – 2184. Srivastava, V.M., Mall, I.D., Mishra, I.M. Equilibrium modeling of ternary adsorption of metal ions onto rice husk ash. Journal of Chemical and Engineering Data 54 (2009) 705-711. Sheng, P.X., Ting, Y.P., Chen, J.P. Biosorption of heavy metals ions (Pb, Cu and Cd) from aqueous solutions by the marine alga sargassum s.p. in single- and multiple-metal systems. Industrial and Engineering Chemistry Research 46 (2007) 2438-2444. Mohan, D., Pittman Jr., C.U., Steele, P.H. Single, binary and multi-component adsorption of copper and cadmium from aqueous solutions on Kraft lignin-a biosorbent. Journal of Colloid and Interface Science 297 (2006) 489-504. Qin, F., Wen, B., Shan, X.Q., Xie, Y.N., Liu, T., Zhang, S.Z., Khan, S.U. Mechanisms of competitive adsorption of Pb, Cu and Cd on peat. Environmental Pollution 144 (2006) 669-680. Srivastava, V.C., Mall, I.D., Mishra, I.M. Equilibrium modeling of single and binary adsorption of cadmium and nickel onto bagasse fly ash. Chemical Engineering Journal 117 (2006) 79-91. Papageorgiou, S.K., Kouvelos, E.P., Katsaros, F.K. Calcium alginate beads from Laminaria digitata for the removal of Cu+2 and Cd+2 from dilute aqueous metal solutions. Desalination 224 (2008) 293-306. Papageorgiou, S.K., Katsaros, F.K., Kouvelos, E.P., Kanellopoulos, N.K. Prediction of binary adsorption isotherms of Cu+2, Cd+2 and Pb+2 on calcium alginate beads from single adsorption data. Journal of Hazardous Materials 162 (2009) 1347-1354.

Water Production and Wastewater Treatment, Nova Science Publishers, Incorporated, 2010. ProQuest Ebook Central,

Copyright © 2010. Nova Science Publishers, Incorporated. All rights reserved.

Multicomponent Removal of Heavy Metals…

97

[17] Srivastava, V.C., Mall, I.D., Mishra, I.M. Removal of cadmium (II) and zinc (II) metal ions from binary aqueous solution by rice husk ash. Colloids and Surfaces A: Physicochemical and Engineering Aspects 312 (2008) 172-184. [18] Srivastava, V.C., Mall, I.D., Mishra, I.M. Competitive adsorption of cadmium(II) and nickel(II) metal ions from aqueous solution onto rice husk ash. Chemical Engineering and Processing: Process Intensification 48 (2009) 370-379. [19] Toscano, G., Caristi, C., Cimino, G. Sorption of heavy metal from aqueous solution by volcanic ash. Comptes Rendus Chimie 11 (2008) 765-771. [20] Ko, D.C.K., Cheung, C.W., Choy, K.K.H., Porter, J.F., McKay, G. Sorption equilibria of metal ions on bone char. Chemosphere 54 (2004) 273-281. [21] Ko, D.C.K., Porter, J.F., McKay, G. Application of the concentration-dependent surface diffusion model on the multicomponent fixed-bed adsorption systems. Chemical Engineering Science 60 (2005) 5472-5479. [22] Cheung, C.W., Ko, D.C.K., Porter, J.F., McKay, G. Binary metal sorption on bone char mass transport model using IAST. Langmuir 19 (2003) 4144-4153. [23] Cheung, C.W., Choy, K.K.H., Porter, J.F., G. McKay, G. Empirical multicomponent equilibrium and film-pore model for the sorption of copper, cadmium and zinc onto bone char. Adsorption 11 (2005) 15-29. [24] Al-Asheh, S., Banat, F., Al-Rousan, D. Adsorption of copper, zinc and nickel ions from single and binary metal ion mixtures on to chicken feathers. Adsorption Science and Technology 20 (2002) 849-864. [25] Kar, P., Misra, M. Use of keratin fiber for separation of heavy metals from water. Journal of Chemical Technology and Biotechnology 79 (2004) 1313-1319. [26] de la Rosa, G., Reynel-Avila, H.E., Bonilla-Petriciolet, A., Cano-Rodríguez, I., Velasco-Santos, C., Martínez-Hernández, A. Recycling poultry feathers for Pb removal from wastewater: kinetic and equilibrium studies. International Journal of Chemical and Biomolecular Engineering 1 (2008) 185-193. [27] Juang, R.S., Shao, H.J. A simplified equilibrium model for sorption of heavy metal ions from aqueous solutions on chitosan. Water Research 36 (2002) 2999-3008. [28] Vieira, R.S., Guibal, E., Silva, E.A., Beppu, M.M. Adsorption and desorption of binary mixtures of copper and mercury ions on natural and crosslinked chitosan membranes. Adsorption 13 (2007) 603-611. [29] Kaminski, W., Tomczak, E., Jaros, K. Interactions of metal ions sorbed on chitosan beads. Desalination 218 (2008) 281-286. [30] Kim, D.S. The removal by crab shell of mixed heavy metal ions in aqueous solution. Bioresource Technology 87 (2003) 355-357. [31] Sprynskyy, M., Buszewski, B., Terzyk, A.P., Namiesnik, J. Study of the selection mechanism of heavy metal (Pb+2, Cu+2, Ni+2 and Cd+2) adsorption on clinoptilolite. Journal of Colloid and Interface Science 304 (2006) 21-28. [32] Prasad, M., Xu, H., Saxena, S. Multi-component sorption of Pb(II), Cu(II) and Zn(II) onto low-cost mineral adsorbent. Journal of Hazardous Materials 154 (2008) 221-229. [33] Boujelben, N., Bouzid, J., Elouear, Z. Adsorption of nickel and copper onto natural iron oxide-coated sand from aqueous solutions: study in single and binary systems. Journal of Hazardous Materials 163 (2009) 376-382.

Water Production and Wastewater Treatment, Nova Science Publishers, Incorporated, 2010. ProQuest Ebook Central,

Copyright © 2010. Nova Science Publishers, Incorporated. All rights reserved.

98

H. E. Reynel-Avila, D. I. Mendoza-Castillo, V. Hernández-Montoya et al.

[34] Pagnanelli, F., Mainelli, S., De Angelis, S., Toro, L. Biosorption of protons and heavy metals onto olive pomace: modelling of competition effects. Water Research 39 (2005) 1639-1651. [35] Martinez-Garcia, G., Bachmann, R.T., Williams, C.J., Burgoyne, A., Edyvean, R.G.J. Olive oil waste as a biosorbent for heavy metals. International Biodeterioration and Biodegradation 58 (2006) 231-238. [36] Fiol, N., Villaescusa, I., Martínez, M., Miralles, N., Poch, J., Serarols, J. Sorption of Pb(II), Ni(II), Cu(II) and Cd(II) from aqueous solution by olive stone waste. Separation and Purification Technology 50 (2006) 132-140. [37] Pérez-Marín, A.B., Ballester, A., González, F., Blázquez, M.L., Muñoz, J.A., Sáez, J., Meseguer-Zapata, V. Study of cadmium, zinc and lead biosorption by orange wastes using the subsequent addition method. Bioresource Technology 99 (2008) 8101-8106. [38] Ma, W., Tobin, J.M. Development of multimetal binding model and application to binary metal biosorption onto peat biomass. Water Research 37 (2003) 3967-3977. [39] McKay, G., Porter, J.F. Equilibrium parameters for the sorption of copper, cadmium and zinc ions onto peat. Journal of Chemical Technology and Biotechnology 69 (1997) 309-320. [40] Balasubramanian, R., Perumal, S.V., Vijayaraghavan, K. Equilibrium isotherm studies for the multicomponent adsorption of lead, zinc, and cadmium onto indonesian peat. Industrial and Engineering Chemistry Research 48 (2009) 2093-2099. [41] Ho, Y.S., Wase, D.A.J., Forster, C.F. Kinetic studies of competitive heavy metal adsorption by sphagnum moss peat. Environmental Technology 17 (1996) 71-77. [42] Ho, Y.S., McKay, G. Competitive sorption of copper and nickel ions from aqueous solution using peat. Adsorption 5 (1999) 409-417. [43] Liu, Z., Zhou, L., Wei, P., Zeng, K., Wen, C., Lan, H. Competitive adsorption of heavy metal ions on peat. Journal of China University of Mining and Technology 18 (2008) 255-260. [44] Qin, F., Wen, B. Single- and multi-component adsorption of Pb, Cu and Cd on peat. Bulletin of Environmental Contamination and Toxicology 78 (2007) 265-269. [45] Al-Asheh, S., Duvnjak, Z. Sorption of cadmium and other heavy metals by pine bark. Journal of Hazardous Materials 56 (1997) 35-51. [46] Otero, M., Rozada, F., Morán, A., Calvo, L.F., García, A.I. Removal of heavy metals from aqueous solution by sewage sludge based sorbents: competitive effects. Desalination 239 (2009) 46-57. [47] Pagnanelli, F., Esposito, A., Veglio, F. Multi-metallic modelling of biosorption of binary systems. Water Research 36 (2002) 4095-4105. [48] Ahmadpour, A., Wang, K., Do, D.D. Comparison of models on the prediction of binary equilibrium data of activated carbons. AIChE Journal 44 (1998) 740-752. [49] Kundu, S., Gupta, A.K. Arsenic adsorption onto iron oxide-coated cement (IOCC): regression analysis of equilibrium data with several isotherm models and their optimization. Chemical Engineering Journal 122 (2006) 93-106. [50] Foo, K.Y., Hameed, B.H. Insights into the modeling of adsorption isotherm systems. Chemical Engineering Journal 156 (2010) 2-10. [51] Kumar, K.V., Sivanesan, S. Comparison of linear and non-linear model in estimating the sorption isotherm parameters for safranin onto activated carbon. Journal of Hazardous Materials B123 (2005) 288-292.

Water Production and Wastewater Treatment, Nova Science Publishers, Incorporated, 2010. ProQuest Ebook Central,

Multicomponent Removal of Heavy Metals…

99

Copyright © 2010. Nova Science Publishers, Incorporated. All rights reserved.

[52] Bonilla-Petriciolet, A., Lira-Padilla, M.G., Soto-Becerra, C. Aplicación del método de recocido simulado en la regresión de isotermas de adsorción. Revista Internacional de Contaminación Ambiental 21 (2005) 201-206. [53] Matott, L.S., Rabideau, A.J. ISOFIT – A program for fitting sorption isotherms to experimental data. Environmental Modelling and Software 23 (2008) 670 – 676. [54] Sheindorf, C., Rebhum, M., Sheintuch, M. A Freundlich-type multicomponent isotherm. Journal of Colloid Interface Science 79 (1981) 136-142.

Water Production and Wastewater Treatment, Nova Science Publishers, Incorporated, 2010. ProQuest Ebook Central,

Copyright © 2010. Nova Science Publishers, Incorporated. All rights reserved. Water Production and Wastewater Treatment, Nova Science Publishers, Incorporated, 2010. ProQuest Ebook Central,

In: Water Production and Wastewater Treatment Editors: B. Antizar-Ladislao and R. Sheikholeslami

ISBN 978-1-61728-503-5 © 2011 Nova Science Publishers, Inc.

Chapter 6

THE ADSORPTION OF DYES ON WASTE TYRE DERIVED ACTIVATED CARBON O.S. Chan, C.W. Wong and G. McKay Department of Chemical and Biomolecular Engineering, Hong Kong University of Science and Technology, Clear Water Bay, Kowloon, Hong Kong

Copyright © 2010. Nova Science Publishers, Incorporated. All rights reserved.

ABSTRACT In this chapter, waste tyre was used as a raw material for the production of activated carbons through pyrolysis. Tyre char (TC) and two tyre derived activated carbons (TAC1 and TAC2) were produced. The TAC1 was activated at 950°C for 4 hr in 85% CO2 and 15% N2 atmosphere while TAC2 was firstly treated with 3M HCl for 18 hr and then pyrolysed at 950°C in steam generated by heating water at 0.45mL/min and 150mL/min of N2 for 4hr. The tyre char and tyre derived activated carbon were characterised with regard to their physical and chemical properties. The adsorption of acid (Acid Blue 25) and basic (Methylene Blue) dyes in aqueous solution were studied. Equilibrium data were fitted to various adsorption isotherms including Langmuir, Freundlich, Redlich-Peterson and Langmuir-Freundlich. It was found that the adsorption capacity of MB was higher than of AB25 for all the adsorbents prepared from tyre. It also showed that the adsorption capacity increased after higher temperature activation and more significantly after acid demineralization treatment. This was due to the fact that tyre char was practically a nonporous material, with a higher temperature treatment, a mesoporous structure was developed. With acid demineralisation, activated carbon with a higher micro- and mesoporosity and an increase in the surface area was further developed.

1. INTRODUCTION It is estimated that the total number of waste tyres in China already reached more than 112 million pieces in 2004 and the amount projected to 2010 will be well over 200 million(Anon, 2005). The disposal of the scrap tyre will be a severe global environmental problem that needs to be solved in the near future as the waste tyres do not decompose easily,

Water Production and Wastewater Treatment, Nova Science Publishers, Incorporated, 2010. ProQuest Ebook Central,

Copyright © 2010. Nova Science Publishers, Incorporated. All rights reserved.

102

O. S. Chan, C. W. Wong and G. McKay

this is due to chemically cross-linked rubber molecules that neither melt nor dissolve. The main hazard of having piles of tyres packing together is the possibility of fire or become a mosquito breeding ground. Therefore, reuse or recycling of waste tyres becomes an important social concern. As a result, the "Regulations on the Management of the Recovery and Reuse of Scrap Tyres" in collaboration with the Environment and Resources Committee of the National People's Congress has been drafted in 2005. The most commonly adopted disposal method for used tyre is to landfill especially since the method of landfilling has been improved continuously due to granulation (Warith and Rao, 2006), but it is still an uneconomical and non-environmental friendly method of disposal as the landfill space become more scarce. It will be a significant improvement if the waste tyres can be recycled and reprocessed into more valuable materials. Pyrolysis of scrap tyre is widely accepted as an alternative process for recovering the value in scrap tyres. However, the process is very energy intensive, even though the tyre derived fuel from pyrolysis of tyre produces same amount of energy as oil and 25-80% more than coal. There will be 30-40weight percent solid tyre char produced during pyrolysis; hence the process economy depends strongly on its application and market value, for example the light oil(Cunliffe and Williams, 1998), tyre char hydrocarbon, reusable adsorbent(Lin et al., 2008) On the other hand, activated carbon has been widely used for the adsorption process in wastewater treatment. With high adsorption surface and pore development, it is capable to adsorb the toxic materials in discharged effluents, for instance, dyestuffs and heavy metals. However, this high value added product is relatively expensive, recently, many researches have been focused on the production of activated carbon like material by using low cost precursors, such as sawdust (Hamadi et al., 2001), cherry stones (Olivares-Marin et al., 2008), peanut shell (Xu and Liu, 2008), bagasse (Valix et al., 2009) and bamboo (Chan et al., 2008). Some studies had showed that the activation of the tyre char could result in a product with reasonable desirable properties such as surface area and porosity similar to the commercially available activated carbon. For example, tyre derived activated carbon have been used in the adsorption of methane and SO2 (Brady et al., 1996), heavy metals, Pb and Hg, adsorption by using tyre derived activated carbon (Alexandre-Franco et al., 2008; Manchon-Vizuete et al., 2005). These recycled materials could achieve an acceptable adsorption capacity since the raw material was relatively inexpensive, so the cost of the whole adsorption process could be greatly reduced. Usually, when preparing the tyre char from the raw material, it will follow two basic steps: raw material preparation and low-temperature carbonisation. For the preparation of tyre derived activated carbon, firstly, it will start with shredding the tyre into the desired size and removal of the steel threads with a magnet. Then, it will undergo low temperature pyrolysis at 450-650˚C in a nitrogen environment to break down the cross-linkage between carbon atoms and eliminate the bulk of the volatile matter to produce tyre char with carbon content of around 30-45% weight percent (Bansal et al., 1988). From the literature, the tyre char itself without further activation can be utilised directly for adsorption processes. However, further improvement of their adsorption potential, i.e. surface area, porosity and surface functional groups, can be obtained through further activation. The methods for producing activated carbons are generally divided into physical and chemical activation. Chemical activation is an activation process that comprises heat treatment of the precursor impregnated with chemical agent such as dehydrating and oxidants,

Water Production and Wastewater Treatment, Nova Science Publishers, Incorporated, 2010. ProQuest Ebook Central,

Copyright © 2010. Nova Science Publishers, Incorporated. All rights reserved.

The Adsorption of Dyes on Waste Tyre Derived Activated Carbon

103

followed by activating the mixture at a temperature between 400-700 oC in a single step(Bansal et al., 1988). There are a wide variety of activating agents such as potassium hydroxide, nitric acid, zinc chloride, ferric iron and sodium carbonate commonly used in commercial production. However, using potassium hydroxide in producing the tyre activated carbons is comparatively the most popular method after some preliminary tests (Teng et al., 2000). Physical activation is another popular activation method for tyre char activation. Usually, it includes two main steps, the carbonisation or pyrolysis of the carbonaceous precursor in an inert atmosphere with nitrogen or helium and then activation by steam, carbon dioxide, air or a mixture of these gases at an elevated temperature above 750oC. Tyre char is further activated for developing the internal porous structure in the carbonised precursor by reacting with the oxidising agents to obtain larger specific surface area and porosity. More recently, some researchers work on the demineralisation and found out that this process can decrease the inorganic impurities as well as to reduce the amount of undesired ash content. Typically, acid was used to remove minerals, in the form of mineral ions. It is generally carried out prior to the physical activation, the most commonly used pre-treatment agents are HCl, H2SO4, HNO3 and HF. The effect of demineralisation is highly dependent on the agent used, the concentration of pre-treatment agent, demineralisation time and char to acid ratio and especially the starting precursor. Shah et al used both HCl and H2SO4 to do the demineralization. After the application of acid treatment of samples, the carbon concentration was significantly increased, indicating the removal of non-carbon phases with acid. Higher carbon concentration was achieved with HCl treatment 93% than H2SO4 around 87%(Shah et al., 2006). This may indicate that the HCl has a better potential for the production of low ash activated carbon. Overall, the chemical and physical natures of physically activated carbons are very dependent on the activation parameters, such as activating agent, activation temperature, time, and demineralisation. In this chapter, two activated carbons from tyre char were prepared; one by physical activation with CO2 only while the other one was first demineralised with acid then physical activated with a mixture of steam and N2. The adsorption of dyes (Acid Blue25, AB25 and Methylene Blue, MB) was also conducted on the produced carbons to determine the dye adsorption capacity of the produced carbon. In additional, the experimental adsorption data were modelled using the conventional isotherm equations include Langmuir, Freundlich, Redlich- Peterson and Langmuir-Freundlich isotherms.

2. MATERIALS AND METHODS 2.1. Materials The starting material was the pyrolysed tyre char, TC, from a local waste tyre processing plant. The tyre char would undergo a series of pre-treatment prior to the activation to ensure all the tyre chars were treated in the same way to give the quality consistence of the tyre char. First, the tyre char would be crushed and sieved to a particle size smaller than 2mm, then the steel would be removed by the magnet. Finally, the tyre char would be pyrolysed once again at 5500C for 2 hour to ensure the volatile compounds in the char was removed.

Water Production and Wastewater Treatment, Nova Science Publishers, Incorporated, 2010. ProQuest Ebook Central,

104

O. S. Chan, C. W. Wong and G. McKay

2.2. Preparation of Adsorbents TAC1 preparation: Approximately 30g of tyre char was placed onto the ceramic holder and placed into the tube furnace. The activation set up is shown in Figure 1. Nitrogen was used as a purge gas at 500mL/min for 1 hour. The tyre char was heated at 15 0C /min until 9500C. The sample was then activated at the same temperature for 4 hr using 425mL/min of CO2 and 75mL/min of N2. When activation was completed, the sample was cooled to room temperature in the furnace under nitrogen flow. The activated carbon was washed thoroughly with DI water until the pH is close to neutral , then it was dried in an oven at 110 °C for 24hr before use. TAC2 preparation: Approximately 30g of tyre char was put into the glass beaker with 3M HCl at 1:3 char to acid mass ratio for acid demineralisation to remove ash and impurities. The tyre char was demineralised for 18 hr and filtered thoroughly and rinsed with deionised water to remove the residual acid until neutral pH. After drying in an oven at 110 °C for 24hr, the demineralised sample was then placed onto the ceramic holder and heated in the horizontal tube furnace. The activation set up is shown in Figure 1. Nitrogen was used as a purge gas at 500mL/min for 1 hour. The demineralised tyre char was heated at 15 0C /min until 9500C. The sample was then activated at the same temperature for 4 hr with steam generated by heating water at 0.45mL/min, and 150mL/min of N2. When activation was completed, the sample was cooled to room temperature in the furnace under nitrogen flow. The activated carbon was washed thoroughly with DI water until the pH is close to neutral. After drying in an oven at 110 °C for 24hr the sample was ready for use. The carbon yield was calculated from:

Copyright © 2010. Nova Science Publishers, Incorporated. All rights reserved.

Yield= Wac/ Wtyre char x 100

(1)

where Wac is the weight of activated carbon and W tyre char is the weight of total tyre char input to produce activated carbon.

Figure1. Schematic diagram of the activation furnace.

Water Production and Wastewater Treatment, Nova Science Publishers, Incorporated, 2010. ProQuest Ebook Central,

The Adsorption of Dyes on Waste Tyre Derived Activated Carbon

105

2.3. Characterisation of Adsorbents The total surface area of the activated carbon was determined from N2 adsorption isotherm at 77K in a Quantachrome Autosorb 1 CLP. The samples were degassed at 200˚C in vacuum for 12 hours prior to the adsorption measurements. Total surface areas were calculated using the BET equation (Brunauer et al., 1938). The molecular area of the nitrogen adsorbate was taken as 16.2 Å2. The total pore volumes were calculated by converting the nitrogen gas adsorbed at a relative pressure of 0.98 to the volume of liquid adsorbate. The micropore volume and micropore surface area were estimated using the t-plot method(Lippens and de Boer, 1965). The pore size distribution was determined by the application of BJH equation (Barrett et al., 1951). An elemental analyser, model ELEMENTAR VARIO EL III, was used to determine the mass fractions of the carbon, hydrogen, nitrogen and sulphur content (in weight percent) of the waste tyre, tyre char and activated carbon. For each sample, at least three determinations were conducted until the confidence level was within 5%. The average of the data was used as the results.

Copyright © 2010. Nova Science Publishers, Incorporated. All rights reserved.

2.4. Adsorption of Dyes from Solution Adsorbates: Acid Blue 25 (AB25) and Methylene Blue (MB) were supplied from Aldrich Chemical Company, Inc. and used directly without further treatment. Their properties are listed in Table 1: In this chapter, Acid Blue 25 and Methylene Blue were used, this is because typically acid dye is more difficult to biodegraded and Methylene Blue number is widely used as an indicator of the adsorptive capacity of activated carbon. This parameter is related to the macro- and mesopore capacity of activated carbon(Lussier et al., 1994). Equilibrium time determination: The time required to reach equilibrium was determined by using 8 jars containing 50mL of 2mmol/L dye solution and in contact with 0.05g of activated carbon. The bottles were sealed and placed in a shaker at 120 rpm. Every 3 days, one of the bottle was removed from the shaker and samples were then withdrawn by a syringe, filtered through a 0.22 μm syringe filter and diluted to the appropriate level and liquid-phase dye concentration,Ct, was then determined by measuring the absorbance of the samples at maximum absorbance (λmax) which is 601 nm for Acid Blue 25 (AB 25) and 661 nm for Methylene Blue (MB) with a Varian Cary 1E UV/Vis spectrophotometer. Table 1. Dye properties

Abbreviation Molecular weight (g/mol) Dye content (%)

Acid Blue 25 AB25 416.4 45

Methylene Blue MB 319.9 95

Maximum wavelength (nm)

601

664

Charge

-1

1

Molecular size (Ǻ) (WxLxT)

12.45x10.17x2.53

7x13.1x2.12

Water Production and Wastewater Treatment, Nova Science Publishers, Incorporated, 2010. ProQuest Ebook Central,

O. S. Chan, C. W. Wong and G. McKay

106

Figure2. Structure of Acid Blue 25 and Methylene blue.

Copyright © 2010. Nova Science Publishers, Incorporated. All rights reserved.

The liquid-phase dye concentration was measured at different time intervals and equilibrium was reached when the adsorption capacity of the adsorbent remains constant. Equilibrium adsorption tests: A constant mass of adsorbent (0.05 g) was weighed into 75 mL glass bottles and in contacted with 50 mL of dye solutions of different initial concentration. The bottles were sealed and placed in a shaker at 120 rpm until equilibrium was reached. Samples were then withdrawn by a syringe, filtered through 0.22 μm syringe filter and diluted to the appropriate level and analysed using a Varian Cary 1E UV/Vis spectrophotometer to determine the residual equilibrium liquid-phase dye concentration, Ce. The equilibrium adsorption capacity, qe (mmol/g), at different dye concentrations was determined by a mass balance on the dye:

qe = (Co − Ce )

V m

(2)

where C0 (mmol/l) is the initial concentration, Ce (mmol/l) is the equilibrium concentration in the liquid phase, V is the volume of liquid phase (L), and m is the mass of the absorbent (g). The plot of equilibrium adsorption capacity against equilibrium concentration in the liquid phase graphically depicts the equilibrium isotherm.

3. RESULTS AND DISCUSSION 3.1. Activated Carbon Characterization Table 2 shows the properties of the tyre char and two tyre derived activated carbons. The yield of the carbon dioxide activated carbon was a lot higher than the steam activated carbon. This is possibly due to the differences of activation energy between the steam and carbon dioxide. The activation reactions are as follows: For carbon dioxide as an activating agent,

Water Production and Wastewater Treatment, Nova Science Publishers, Incorporated, 2010. ProQuest Ebook Central,

The Adsorption of Dyes on Waste Tyre Derived Activated Carbon

107

C + CO2 → 2CO ΔH = +159kJ/mol For steam as an activating agent, C + H2O → CO + H2 ΔH = +117kJ/mol

Copyright © 2010. Nova Science Publishers, Incorporated. All rights reserved.

CO + H2O → CO2 + H2 ΔH = -41kJ/mol The activation energies of the reactions between char and carbon dioxide or steam are 159 kJ/mol and 117kJ/mol that the reaction of carbon dioxide requires higher activated energy(Marsh and Rodriguez-Reinoso, 2006) . Also, carbon dioxide has a greater molecular size which diffuses slower and with more difficulty into the narrow pores than steam. Another reason may due to the greater stability of activated char produced by carbon dioxide and remained longer on the char surface, and results in a slower reaction rate. As a result, the burnoff will be significantly decreased for the carbon dioxide activated carbon. Another possible reason will be that for the steam activation, the reaction of first equation will give out carbon monoxide, which will then react with H2O to form CO2, the the CO2 will react with carbon and consume more carbon. In this case, the carbon will go through double activation and hence higher burn-off. It is also observed by other researchers that the carbon dioxide activated carbon with a burnoff 75% of that produced by steam(Li et al., 2005). Another group of researchers found out that the tyre chars possess higher reactivity with steam activation than carbon dioxide and found that the burnoff by carbon dioxide was 69.5% of that produced by steam on average while steam activated carbon had a higher BET surface area than carbon dioxide activation (Gonzaez et al., 2006). The BET surface area and pore volumes are also shown in the Table 2 for the tyre char and two produced carbon, TC, TAC1 and TAC2, respectively. For the starting material, tyre char, there were no micropores and only limited amount of mesopores was observed. There was no micropores development in TAC1 as well, however, the carbon dioxide activation allowed mesopores to develop and the pore volume was three times higher compared to the original tyre char. Micropores and a significant amount of mesopores were developed with acid demineralisation and steam activation. TAC2 has shown a relatively high surface area that is comparable to the commercial activated carbon (Calgon F400 with BET surface area around 800m2/g). Table 2. Properties of the adsorbents

Name

Yield (%)

Surface Area (m2/g)

Vmic (cm3/g)

Vmes (cm3/g)

Vtot (cm3/g)

C (%)

H (%)

N (%)

S (%)

O(%)

TC

--

41.92

0

0.11

0.11

81.9

1.39

2.4

3.16

11.15

TAC1

59.2

186.8

0

0.32

0.32

74.6

1.82

0.28

3.85

19.45

TAC2

28.6

970

0.118

1.122

1.24

69.3

1.57

0.07

0.39

28.67

The influence of the two activating agents on the pore structure development of the activated carbon is different. Generally, the carbon activated by steam at high burn off resulted in a wider pore size distribution for both meso- and macropores (Arriagada et al., Water Production and Wastewater Treatment, Nova Science Publishers, Incorporated, 2010. ProQuest Ebook Central,

Copyright © 2010. Nova Science Publishers, Incorporated. All rights reserved.

108

O. S. Chan, C. W. Wong and G. McKay

1994; Bansal et al., 1988). It supported the results in this chapter, with a greater burnoff of the TAC2 and with significant amount of mesopores development. Some researchers found that activation with carbon dioxide at a temperature higher than 1000oC resulted in the rapid burn off of the exterior of the particles and opened and widened microporosity (Rodriguez-Reinoso et al., 1995). Increased porosity was accompanied by a parallel increase in surface area. Most of the studies showed that steam was more effective than carbon dioxide as an activation agent. Steam activation not only produced a slightly narrower microporous activated carbon but also could develop the same porosity as carbon dioxide activation with a 50oC less in activation temperature. Some researchers agreed that steam activated carbon gave a better BET surface area than carbon dioxide activated carbon and having a BET surface area in excess of 1000m2/g(Ariyadejwanich et al., 2003; Brady et al., 1996; Gonzalez et al., 2006; San Miguel et al., 2002; San Miguel et al., 2003). Other researchers suggested that the mild oxidation of char with superheated steam resulted in a BET surface area enhancement and microporosity development (Merchant and Petrich, 1992). It was been suggested that the activated carbons obtained by steam activation process were mainly mesoporous with limited microporosity (Lopez et al., 2009). Demineralisation is another possible cause in the significant increase of surface area of TAC2. Although there is limited research on the demineralisation of tyre char activated carbon, possibly because of the high cost in the preparation and intensive washing required after the demineralisation. All of the studies indicated that there were significant improvements of textural properties in the tyre char activated carbon. After acid demineralisation, there was a decrease in burn off, lowering of the ash content and an increased in surface area and porosity. It was found that the demineralised tyre chars had a lower reactivity when activated; this might be due to the catalytic effect of the inorganic compounds such as Ca, Zn, P, etc on the activation. It was also found that the removal of undesired inorganic impurities that blocked the pore structure of the pyrolysed char occurred, leading to an increase of surface area, micro- and mesopore volumes(Ariyadejwanich et al., 2003; Cunliffe and Williams, 1999; Galvagno et al., 2009; Mui et al., 2009; Shah et al., 2006; Ucar et al., 2005; Zhu et al., 2009). This result was extremely positive, since the impurities in the tyre char, such as inorganic oxides and sulphur content might cause a pollution problem and limit its applications could be reduced. Table 2 also shows the elemental analysis results of the prepared carbons. The carbon content of the tyre char is reasonably high with around 82 wt% with small amounts of hydrogen, nitrogen and sulphur. Carbon levels decreased during the activation for both carbon dioxide activation and steam activation. The hydrogen content remained at similar level, while most of the nitrogen was eliminated during the activation. The sulphur seemed remain the same level with the carbon dioxide activation but was almost eliminated in the steam activated carbon with demineralisation. This is because the acid demineralisation will remove the ash-forming minerals, sulphate, pyritic sulfur, and organic sulfur (Mukherjee and Borthakur, 2003). There was significant increase in the oxygen content (by difference method). This might be caused by the oxygen in the steam or CO2 reacting with parts of the carbonised materials to produce CO and CO2, which opened up blocked pores and creating new pores in the carbon (Bansal et al., 1988). As well as this, the reaction of the CO2, steam, air and diluted oxygen formed chemisorbed oxygen, called surface oxygen complexes which possed a broad range of chemical functionality surface group including carboxylic, phenolic and etc, those functional group will be particularly good for adsorbing positive charge molecules.

Water Production and Wastewater Treatment, Nova Science Publishers, Incorporated, 2010. ProQuest Ebook Central,

The T Adsorptionn of Dyes on Waste W Tyre Deerived Activatted Carbon

109

3.2. Dye Adssorption

Copyright © 2010. Nova Science Publishers, Incorporated. All rights reserved.

ption capacityy was directlyy affected byy: Firstly, thee physical andd chemical The adsorp naature of the adsorbent, a succh as pore strructure, chemiical functionaal groups; Seccondly, the naature of the ad dsorbate, its pK Ka, functionall groups preseent, polarity, molecular m dimeension; and finally the expeerimental conditions such as a pH, temperrature and the adsorbate conncentration Haghseresht ett al., 2002). (H As shown in Figures 3 and a 4, the experimental adsorption capacity of the TAC C2 was the hiighest among the three carbbons with aroound 1.15mmool/g of AB25 and around 1.25mmol/g off MB, followeed by the TAC C1 and with very v little adsoorption was obbserved for thhe TC. This w because th was he higher surfaace area and porosity p of thee TAC2 providding more acttive sites to addsorb the dyee molecules. As both dyees were relativvely small molecules m (Tabble 1), the addsorption was mainly takenn place in bothh meso and micropores. As the t total pore volume for TAC2 was the highest, nearlly 4 times highher than TAC1 and 10 timees of TC, hencce, it would bee able to adso orb more of both b dyes. Its adsorption caapacity for AB B25 was higher than the coommercial acttivated carbonn, F400, with only 0.7mmool/g(Chan et al., a 2008). Sim milarly, the addsorption capaacity for the basic b dye was slightly higheer than the aciid dye for all the t carbon. This might be due to two caauses, firstly, the t dimensionn of Methylenne Blue was sm maller than thhe Acid Blue 25, so it coulld diffuse fastter and into thhe smaller poore of the inneer area and voolume of the activated a carbon particle. Seecondly, as thhere was an inccrease in oxyggen context off the activated carbon, thee surface oxyggen complexees that wouldd be more favvourable to addsorb the positively chargedd molecules, i..e. the basic dyye.

Fiigure 3. The AB B25 adsorption isotherm on tyrre char and tyree derived activatted carbon..

Fiigure 4. The MB B adsorption isootherm on tyre char and tyre derived activatedd carbon. Water Production and Wastewater Treatment, Nova Science Publishers, Incorporated, 2010. ProQuest Ebook Central,

O. S. Chan, C. W. Wong and G. McKay

110

However, many researchers used the monolayer capacity that calculated from Langmuir isotherm to compare the adsorption capacity in Table 2, in this case the adsorption of AB25 is more than the MB, this shown that the TAC2 has heterogeneous surface, This indicated that with different degrees of activation, there were some changes in the surface chemistry and functional groups, hence further studies should be carried out to determine the effect of the activation conditions on the surface properties and its effect on dye adsorption.

3.3. Equilibrium Isotherm Modelling The experimental data were fitted into Langmuir, Freundlich, Redlich-Peterson (RP) and Langmuir-Freundlich (LF) equations to determine which isotherm gave the best correlation to the experimental data shown in Figures5, 6and 7. Langmuir isotherm: The Langmuir adsorption isotherm is most widely used for modelling the adsorption behaviour of the adsorbates from a liquid solution(Langmuir, 1916). It has several critical assumptions, which include monolayer adsorption, all sites are identical and energetically equivalent, one adsorbate occupies one site, and the adsorption energy is constant throughout the adsorption process, no interaction between adsorbates in the adjacent sites and finally it has a saturation or equilibrium value, where no more adsorption can take place. Its equilibrium capacity can be represented by the expression:

Copyright © 2010. Nova Science Publishers, Incorporated. All rights reserved.

qe=KLCe/(1+aLCe)

(3)

where qe is the solid phase sorbate concentration at equilibrium (mmol/g), Ce is the aqueous phase sorbate concentration at equilibrium (mmol/L), KL is Langmuir isotherm constant (L/g), aL is Langmuir isotherm constant (L/mmol). The monolayer capacity, qmono, can then be calculated by the following equation: qmono= KL/aL

(4)

Freundlich isotherm: The Freundlich model (Freundlich, 1906) is an empirical model incorporating the heterogeneity of the adsorption energies on the surface. It is commonly used for the description of multilayer adsorption with interaction between adsorbed molecules. The model predicts that the dye concentrations on the material will increase when there is an increase in concentration of the dye in solution. 1

qe = K F C e n

(5)

where qe is the solid phase sorbate concentration in equilibrium (mmol/g), Ce is liquid phase sorbate concentration in equilibrium (mmol/L), KF is Freundlich constant (L/g) and 1/n is the heterogeneity factor. Redlich–Peterson isotherm: Redlich and Peterson have proposed an empirical equation incorporating three parameters. It can be used to represent adsorption equilibria over a wide concentration range, and can be applied to both homogeneous and heterogeneous systems (Redlich and Peterson, 1959). This isotherm combines elements from both the Langmuir and

Water Production and Wastewater Treatment, Nova Science Publishers, Incorporated, 2010. ProQuest Ebook Central,

The Adsorption of Dyes on Waste Tyre Derived Activated Carbon

111

Freundlich equations, and the adsorption mechanism does not follow ideal monolayer adsorption. It reduces to Henry's equation when bR = 0 and to Langmuir isotherm when bR = 1. qe=KRCe/(1+aRCe bR )

(6)

where qe is solid phase sorbate concentration in equilibrium (mmol/g), Ce is liquid phase sorbate concentration in equilibrium (mmol/l), KR is Redlich– Peterson isotherm constant (L/g), aR is Redlich– Peterson isotherm constant and bR is the exponent which lies between 1 and 0. Langmuir-Freundlich isotherm: The Langmuir-Freundlich or Sips isotherm combines the Langmuir and Freundlich equations with three parameters KLF, aLF and bLF. At low sorbate concentrations it effectively reduces to the Freundlich isotherm and thus does not obey Henry's law. At high sorbate concentrations, it predicts a monolayer sorption capacity characteristic of the Langmuir isotherm(Sips, 1948) . qe= KLF Ce1/bLF/(1+aLF Ce1/bLF)

(7)

Copyright © 2010. Nova Science Publishers, Incorporated. All rights reserved.

where qe is the adsorbed amount at equilibrium (mmol/g), where aLF is the LF model isotherm constant (L/g); KLF the LF constant (L/mmolg) and 1/b the LF model exponent. Modelling comparison: Three sets of modelling data of the individual adsorbents are plotted in Figures 3 to 5 and the individual dye adsorption model parameters are shown in Table 3. The sum of the squares of the errors (SSE) is used to identify the best fit model. SSE is chosen as the error anaylysis method is because we are more interested in the monolayer capacity, so the bias toward the data obtained at high end of concentration is desired. SSE= (qe,calc − qe,exp)2

(8)

From Table 3, lowest SSE values were obtained using the Freundlich model for the adsorption of AB25 on TC and TAC1 and for TAC2, the Freundlich model provide thesecond best correlation. The Freundlich model fitted the experimental data of MB better than the other models since it provided the smallest SSE for TC and TAC1, for TAC2 the LangmuirFreundlich model was best followed by Freundlich. Freundlich isotherm could be used to described the adsorption of the dyes onto the tyre carbons, since in this case as less than 5% difference. A two parameter model is better than three parameter.

Water Production and Wastewater Treatment, Nova Science Publishers, Incorporated, 2010. ProQuest Ebook Central,

Water Production and Wastewater Treatment, Nova Science Publishers, Incorporated, 2010. ProQuest Ebook Central,

AB 25

Copyright © 2010. Nova Science Publishers, Incorporated. All rights reserved.

Table 3. Modelling parameters for the dye –carbon system

Langmuir

TC TA C1 TA C2

KL 4.12E01 1.13E +03 2.11E +03

MB

Langmuir

TC TA C1 TA C2

KL 1.13E +00 1.75E +03 2.97E +03

Freundlich aL 5.73E +00 6.98E +03 1.91E +03

SSE 1.56E04 1.30E02 1.00E +00

Qmon o 7.19E02 1.62E01 1.10E +00

KF 7.24E01 2.14E01 1.13E +00

1/n 1.24E +00 1.11E01 3.99E02

Redlich-Peterson SSE 6.24E05 3.65E04 1.01E +00

Freundlich aL 6.02E +00 7.70E +03 2.90E +03

SSE 5.38E04 3.65E02 3.76E01

Qmon o 1.88E01 2.27E01 1.02E +00

KF 2.27E01 2.61E01 1.16E +00

1/n 5.51E01 9.38E02 1.52E01

KR 2.92E +02 4.59E +04 2.23E +03

aR 7.09E +02 2.16E +05 2.01E +03

Langmuir-Freundlich bR 1.00E03 8.91E01 9.97E01

SSE 1.56E04 7.72E03 1.00E +00

Redlich-Peterson SSE 9.62E04 2.39E02 2.60E01

KR 1.13E +00 1.35E +03 1.54E +04

aR 6.02E +00 5.51E +03 1.36E +04

KLF 2.74E02 2.14E01 1.62E +02

bLF 1.00E +00 9.00E +00 1.64E +00

aLF 1.33E +00 1.00E02 1.44E +02

SSE 3.04E03 7.74E03 1.00E +00

Langmuir-Freundlich bR 1.00E +00 9.70E01 8.81E01

SSE 5.38E04 3.63E02 2.43E01

KLF 2.82E +00 3.72E01 5.67E +00

bLF 7.66E01 8.48E +00 3.23E +00

aLF 1.83E +01 4.42E01 4.08E +00

SSE 4.70E04 2.39E02 2.24E01

The T Adsorptionn of Dyes on Waste W Tyre Deerived Activatted Carbon

Copyright © 2010. Nova Science Publishers, Incorporated. All rights reserved.

Fiigure.5. Plot of sorption of dyees on TC with modelling. m

Fiigure.6. Plot of sorption of dyees on TAC1 witth modelling.

Fiigure.7. Plot of sorption of dyees on TAC2 witth modelling. Water Production and Wastewater Treatment, Nova Science Publishers, Incorporated, 2010. ProQuest Ebook Central,

113

114

O. S. Chan, C. W. Wong and G. McKay

4. CONCLUSIONS The preparation of high BET surface area activated carbons from waste tyre by pyrolysis at high temperature with different activating agent had been demonstrated. The high surface area carbon showed nearly three to five times higher adsorption capacity than the original tyre char. The tyre char showed a relatively poor adsorption for both dyes. Both textural properties and chemical properties of the carbon played an important role in the adsorption of both acid and basic dyes. The Freundlich isotherm model can be used to describe the adsorption of the dyes onto the tyre char and tyre derived activated carbon reasonably well.

Copyright © 2010. Nova Science Publishers, Incorporated. All rights reserved.

REFERENCES Alexandre-Franco, M., Fernandez-Gonzalez, C., MacIas-Garcia, A. and Gomez-Serrano, V., Adsorption 2008, 14, (4-5), 591-600. Anon, China Chemical Reporter 2005, (23). Ariyadejwanich, P., Tanthapanichakoon, W., Nakagawa, K., Mukai, S. R. and Tamon, H., Carbon 2003, 41, (1), 157-164. Arriagada, R., Garcia, R. and Reyes, P., J. Chem. Technol. Biotechnol. 1994, 60, (4), 427435. Bansal, R. P., Donnet, J. and Stoeckli, F., Active Carbon. (Marcel Dekker, New York, 1988). Barrett, E. P., Joyner, L. G. and Halenda, P. P., Journal of the American Chemical Society 1951, 73, (1), 373-380. Brady, T. A., Rostam-Abadi, M. and Rood, M. J., Gas Separation and Purification 1996, 10, (2), 97-102. Brunauer, S., Emmett, P. H. and Teller, E., Journal of the American Chemical Society 1938, 60, (2), 309-319. Chan, L. S., Cheung, W. H. and McKay, G., Desalination 2008, 218, (1-3), 304-312. Cunliffe, A. M. and Williams, P. T., Journal of Analytical and Applied Pyrolysis 1998, 44, (2), 131-152. Cunliffe, A. M. and Williams, P. T., Energy and Fuels 1999, 13, (1), 166-175. Freundlich, H. M. F., Zeitschrift Fur Physikalische Chemie Stochiometrie und Verwandtschaftslehre 1906, 57, (4), 385–471. Galvagno, S., Casciaro, G., Casu, S., Martino, M., Mingazzini, C., Russo, A. and Portofino, S., Waste Management 2009, 29, (2), 678-689. Gonzaez, J. F., Encinar, J. M., Gonzaez-Garcia, C. M., Sabio, E., Ramiro, A., Canito, J. L. and Ga an, J., Applied Surface Science 2006, 252, (17), 5999-6004. Gonzalez, J. F., Encinar, J. M., Gonzalez-Garcia, C. M., Sabio, E., Ramiro, A., Canito, J. L. and Ganan, J., Applied Surface Science 2006, 252, (17), 5999-6004. Haghseresht, F., Nouri, S., Finnerty, J. J. and Lu, G. Q., Journal of Physical Chemistry B 2002, 106, (42), 10935-10943. Hamadi, N. K., Xiao Dong, C., Farid, M. M. and Lu, M. G. Q., Chemical Engineering Journal 2001, 84, (2), 95-105. Langmuir, I., The Journal of the American Chemical Society 1916, 38, (2), 2221-2295.

Water Production and Wastewater Treatment, Nova Science Publishers, Incorporated, 2010. ProQuest Ebook Central,

Copyright © 2010. Nova Science Publishers, Incorporated. All rights reserved.

The Adsorption of Dyes on Waste Tyre Derived Activated Carbon

115

Li, S. Q., Yao, Q., Wen, S. E., Chi, Y. and Yan, J. H., Journal of the Air and Waste Management Association 2005, 55, (9), 1315-1326. Lin, C., Huang, C. L. and Shern, C. C., Resources, Conservation and Recycling 2008, 52, (10), 1162-1166. Lippens, B. C. and de Boer, J. H., Journal of Catalysis 1965, 4, (3), 319-323. Lopez, G., Olazar, M., Artetxe, M., Amutio, M., Elordi, G. and Bilbao, J., Journal of Analytical and Applied Pyrolysis 2009, 85, (1-2), 539-543. Lussier, M. G., Shull, J. C. and Miller, D. J., Carbon 1994, 32, (8), 1493-1498. Manchon-Vizuete, E., MacIas-Garcia, A., Nadal Gisbert, A., Fernandez-Gonzalez, C. and Gomez-Serrano, V., J. Hazard. Mater. 2005, 119, (1-3), 231-238. Marsh, H. and Rodriguez-Reinoso, F., Activated Carbon 2006. Merchant, A. A. and Petrich, M. A., Chemical Engineering Communications 1992, 118, 251263. Mui, E. L. K., Cheung, W. H., Valix, M. and McKay, G., Microporous and Mesoporous Materials 2009. Mukherjee, S. and Borthakur, P. C., Fuel 2003, 82, (7), 783-788. Olivares-Marin, M., Fernandez, J. A., Lazaro, M. J., Fernandez-Gonzalez, C., Macias-Garcia, A., Gomez-Serrano, V., Stoeckli, F. and Centeno, T. A., Materials Chemistry and Physics 2008. Redlich, O. and Peterson, D. L., Journal of Physical Chemistry 1959, 63, (6), 1024. Rodriguez-Reinoso, F., Molina-Sabio, M. and Gonzalez, M. T., Carbon 1995, 33, (1), 15-23. San Miguel, G., Fowler, G. D., Dall'Orso, M. and Sollars, C. J., Journal of Chemical Technology and Biotechnology 2002, 77, (1), 1-8. San Miguel, G., Fowler, G. D. and Sollars, C. J., Carbon 2003, 41, (5), 1009-1016. Shah, J., Jan, M. R., Mabood, F. and Shahid, M., Journal of the Chinese Chemical Society 2006, 53, (5), 1085-1089. Sips, R., The Journal of Chemical Physics 1948, 16, (5), 490-495. Teng, H., Lin, Y. C. and Hsu, L. Y., Journal of the Air and Waste Management Association 2000, 50, (11), 1940-1946. Ucar, S., Karagoz, S., Ozkan, A. R. and Yanik, J., Fuel 2005, 84, (14-15), 1884-1892. Valix, M., Cheung, W. H. and McKay, G., Adsorption 2009, 15, (5-6), 453-459. Warith, M. A. and Rao, S. M., Waste Management 2006, 26, (3), 268-276. Xu, T. and Liu, X., Chinese Journal of Chemical Engineering 2008, 16, (3), 401-406. Zhu, J., Shi, B., Chen, L., Liu, D. and Liang, H., Waste Management and Research 2009, 27, (6), 553-560.

Water Production and Wastewater Treatment, Nova Science Publishers, Incorporated, 2010. ProQuest Ebook Central,

Copyright © 2010. Nova Science Publishers, Incorporated. All rights reserved. Water Production and Wastewater Treatment, Nova Science Publishers, Incorporated, 2010. ProQuest Ebook Central,

In: Water Production and Wastewater Treatment Editors: B. Antizar-Ladislao and R. Sheikholeslami

ISBN 978-1-61728-503-5 © 2011 Nova Science Publishers, Inc.

Chapter 7

ADSORPTION OF BASIC DYES BY ACTIVATED CARBON FROM WASTE BAMBOO L.S. Chan1,2, W.H. Cheung2, S. J. Allen1 and G. McKay2 1

School of Chemistry and Chemical Engineering, Queen’s University Belfast, Belfast, Northern Ireland, UK 2 Department of Chemical and Biomolecular Engineering, Hong Kong University of Science and Technology, Clear Water Bay, Kowloon, Hong Kong

Copyright © 2010. Nova Science Publishers, Incorporated. All rights reserved.

ABSTRACT Bamboo, indigenous to Hong Kong and China, is widely used as scaffolding in local construction and building projects. Over 50,000 tonnes of bamboo scaffolding waste is disposed as landfill waste each year. These wastes can be used as a sustainable raw material for the production of a range of high value added activated carbons for various applications e.g. adsorbents, catalysts or catalyst supports. Super-high surface area activated carbons were produced by thermal activation of waste bamboo scaffolding with phosphoric acid. Surface areas up to 2500m2/g were produced. In order to evaluate the adsorption capacity of the produced carbons, dye adsorption was conducted on the carbons produced and compared with a commercially available carbon. Two basic dyes, namely, Basic Yellow 11 (BY11) and Maxilon Red GRL 200% (MR) were used. It was found that both basic dyes were readily adsorbed onto the produced carbon and were up to three times higher than the commercial carbon. In addition, experimental results were fitted to equilibrium isotherm models including Langmuir, Freundlich, and RedlichPeterson.

INTRODUCTION In the textile industry, dyestuff is used to provide garments with different colour and shade. A consumption rate of approximately one billion kg of dye was reported in 1994 (Marc, 1996). An estimated 10-20% of dyes (active substance) used is lost in residual liquors through exhaustion and washing operations. Reverse osmosis, ion exchange, coagulation, precipitation, catalytic reduction, herbal filtration, electrodialysis and adsorption, all have

Water Production and Wastewater Treatment, Nova Science Publishers, Incorporated, 2010. ProQuest Ebook Central,

Copyright © 2010. Nova Science Publishers, Incorporated. All rights reserved.

118

L. S. Chan, W. H. Cheung, S. J. Allen et al.

been used to treat textile effluents. However, available effluent treatment processes for dye containing effluents are currently only capable of removing about half the dyes lost in residual liquors to the environment. Adsorption is the most widely use and effective treatment process for textile effluent. Numerous adsorbents including inorganic, agricultural and shellfish by-products have been considered for adsorption (Juang, et al., 1996, Namasivayam, et al., 2001, Garg, et al., 2003, Walker, et al., 2003, Ramesh, et al., 2005, Crini, 2006). The use of activated carbons however, has been widely favoured because of their high adsorption capacities and amphoteric properties, which enable their adsorption of both cationic and anionic pollutants in effluents (Corapcioglu and Huang, 1987, Chern and Wu, 1999, Al-Degs, et al., 2000, Annadurai, et al., 2000). The relatively high cost of activated carbon and its regeneration problems hindered its usage, especially in third world countries. A challenge in the field of activated carbon production is to produce specific materials with given properties including pore size distribution and surface area from low cost precursors and at low temperature. In recent years, considerable interest has been focused on low cost alternative materials for the production of active carbons from wastes and agricultural by-products such as waste tyres, fruit stones, oil-palm shell and bagasse (Tsai, et al., 1998, Ahmedna, et al., 2000, Marshall, et al., 2000, Valix, et al., 2004) with no need of regeneration. In Hong Kong and South-East China, with the booming in the construction and building industry, a large quantity of construction wastes is generated each year. One of these wastes is bamboo, which is uniquely used in this region as scaffolding for building projects. Over 50,000 tonnes of bamboo scaffolding each year is dumped as construction waste in Hong Kong. With limiting landfilling space available, alternative usages of this sustainable material are being sought. Bamboo is a tropical plant and is indigenous to Southern Asia, including China, Hong Kong, Thailand and Vietnam. It has a rapid growth rate and consumes little energy (0.5MJ/kg). Bamboo waste can be used as the precursor for the production of a range of activated carbons and carbon chars due to its high carbon content (44%). The chars can be further treated using various chemicals and over a range of temperatures to produce a selection of activated carbons for various uses (Wu, et al., 1999, Ohe, et al., 2003). Bamboobased activated carbon has the potential to be a sustainable and commercially available for the treatment of 1) gaseous pollutants, 2) liquid pollutants in industrial effluents, 3) in drinking water filtration applications and 4) fuel cell and electronic applications. Phosphoric acid activation is a conventional method for the preparation of active carbon from lignocellulosic materials (Laine, et al., 1989). The precursor is impregnated with a solution of phosphoric acid, heat treated to 600˚C and washed with water to remove excess acid. Phosphoric acid induces important changes in the pyrolytic decomposition of the lignocellulosic materials as it promotes depolymerisation, dehydration and redistribution of constituent biopolymers (Jagtoyen and Derbyshire, 1993, 1998). So far, only limited research has been carried out on bamboo as precursor utilizing the high temperature physical activation. BET-nitrogen surface area ranges from 491 to 1038 m2/g have been obtained (Wu, Tseng and Juang, 1999, Abe, et al., 2001, Kannan and Sundaram, 2001, Asada, et al., 2002). In this chapter, the feasibility of producing activated carbon from waste bamboo scaffolding by low temperature chemical activation with phosphoric acid will be investigated. In addition, the adsorption capacities of basic dyes, namely, Basic Yellow 11 (BY11) and Maxilon Red GRL 200% (MR), is conducted on the produced carbon and compared with a commercially available carbon, Calgon F400.

Water Production and Wastewater Treatment, Nova Science Publishers, Incorporated, 2010. ProQuest Ebook Central,

Adsorption of Basic Dyes by Activated Carbon from Waste Bamboo

119

MATERIALS AND METHODOLOGY Materials The received waste bamboo scaffolding was washed with water and reduced to size by hammer milling prior to experiment. Particle size of 500 – 710 μm was used. The average chemical composition of bamboo is measured by an elemental analysis and is shown in Figure 1. The major elements are carbon (47.6%) and oxygen (43.9%) accounting for around 91% of bamboo. Other elements include hydrogen (6.5%), nitrogen (0.3%), sulphur (0.3%) and ash (1.4%). This raw material has been pre-treated by transfer into alumina containers, soaking and saturating with ortho-phosphoric acid (H3PO4) at different acid to bamboo ratio (Xp). The mixture has been stirred thoroughly to ensure homogenous mixing of the bamboo and H3PO4. Then, the samples were subjected to a two-step heating process at 150˚C for two hours and then at 600˚C in a furnace under flowing nitrogen for a range of time. After heating, samples were cooled, washed and dried for further analysis and characterisation.

Carbon Characterisation Test Chemical activated carbons were characterised by BET surface area, pore size distribution, elemental analysis and dye adsorption equilibrium capacity. The apparent surface area of the activated carbon was determined from N2 adsorption at 77K in a Coulter 3100 analyzer using the Brunauer, Emmett and Teller (BET) equation ((Brunauer, et al., 1938). The molecular area of the nitrogen adsorbate was taken as 16.2 Å2.

Copyright © 2010. Nova Science Publishers, Incorporated. All rights reserved.

Sulfur (S) 0.3%

Ash 1.4%

Carbon (C) 47.6%

Oxygen (O) 43.9%

Hydrogen (H) 6.5%

Nitrogen (N) 0.3% Carbon (C)

Hydrogen (H)

Nitrogen (N)

Oxygen (O)

Sulfur (S)

Ash

Figure 1. Elemental composition of bamboo.

The micropore volume was estimated by applying the Horvath Kawazoe (HK) method (Horvath and Kawazoe, 1983) which assumes slit pore shapes on the nitrogen adsorption isotherms. Mesopore volume was estimated using the Kelvin equation (Bansal, et al., 1988). Water Production and Wastewater Treatment, Nova Science Publishers, Incorporated, 2010. ProQuest Ebook Central,

.

L. S. Chan, W. H. Cheung, S. J. Allen et al.

120

The total pore volumes were calculated by converting the nitrogen gas adsorbed at a relative pressure 0.98 to the volume of liquid adsorbate.

Dye Adsorption Equilibrium Capacity Test The basic dye adsorption test was used to determine the adsorption capacity of the produced carbons using two basic dyes; Basic Yellow 11 (BY11) was supplied by SigmaAldrich Chemical Company while Maxilon Red GRL 200% (MR) was supplied by Hong Kong Polytechnic University. Table 1 and Figure 2 show the properties and structure of the dyes, respectively. The dimensions of dye molecules were estimated using the software, ChemSketch by ACD, Inc. A fixed mass of activated carbon, 0.020g was weighed into 80 mL glass bottles and brought into contact with 50mL of dye solution with predetermined initial dye concentrations. The bottles were sealed and agitated continuously at 200 rpm in the thermostatic shaker bath and maintained at a temperature of 25 °C ± 1 °C until equilibrium was reached. At time t = 0 and equilibrium, the dye concentrations of the solutions were measured by Varian Cary 1E UV-Vis Spectrophotometer. These data were used to calculate the adsorption capacity, qe, of the adsorbent. The adsorption capacities (qe) of the each activated carbon were determined by:

Copyright © 2010. Nova Science Publishers, Incorporated. All rights reserved.

qe =

V (C0 − Ce ) m

(1)

where qe = the dye concentration on the adsorbent at equilibrium (mmol/g) C0 = the initial dye concentration in the liquid phase (mmol of dye / L), Ce = the liquid-phase dye concentration at equilibrium (mmol of dye / L), V = the total volume of dye-activated carbon mixture (L), m = mass of activated carbon used (g). Finally, the adsorption capacity, qe, was plotted against the equilibrium concentration, Ce. Table 1. Physical and chemical properties of the basic dyes Basic Yellow 11 (BY11)

Colour Index Abbreviation Molecular Weight (g/mol) Dye Content (%) Chromophore Maximum Wavelength, λmax (nm) Charge Width (Å) Depth (Å) Thickness (Å)

48055 BY11 372.9 20 Methine 412 +1 15.293 14.918 4.148

Water Production and Wastewater Treatment, Nova Science Publishers, Incorporated, 2010. ProQuest Ebook Central,

Maxilon Red GRL 200% (MR) 110825 BR46 501 73 Monoazo 530 +1 13.325 14.569 6.063

Adsorption of Basic Dyes by Activated Carbon from Waste Bamboo

121

Copyright © 2010. Nova Science Publishers, Incorporated. All rights reserved.

Maxilon Red GRL 200%, (MR).

Basic Yellow 11, (BY11). Figure 2. Molecular structure of the basic dyes.

RESULTS AND DISCUSSION Basic Dyes Adsorption Capacities Three bamboo carbons, namely, two carbons with high surface area (HSA1 and HSA2) and one with low surface area (LSA) were prepared for the study of basic dyes adsorption capacities, their properties are shown in Table 2.

Water Production and Wastewater Treatment, Nova Science Publishers, Incorporated, 2010. ProQuest Ebook Central,

L. S. Chan, W. H. Cheung, S. J. Allen et al.

122

Properties

HSA1

HSA2

LSA

F400

BET surface area (m2/g) Micropore area (m2/g) External surface area (m2/g) Average pore diameter (Å) Total pore volume (cc/g) Micropore volume (cc/g)

2471 2172 245 22.20 1.341 1.023

2200 1398 802 34.31 1.887 0.651

758 724 34 22.33 0.423 0.377

747 673 74 25.18 0.470 0.348

Micropore Volume Total Pore Volume

0.763

0.345

0.891

0.740

Both high surface area carbons adsorbed over three times more than the F400 and LSA for both basic dyes as shown in Figures 3 and 4 while F400 and LSA have very similar low capacities, this demonstrates that surface area and porosity also plays an important role in the adsorption of basic dyes. Both HSA1 and HSA2 have similar high adsorption capacities for both basic dye systems. This may also link to the similar molecule size of the basic dyes that can penetrate further into activated carbon porous structure with little steric hindrance. Overall, the produced bamboo carbons (HSA1 and HSA2) adsorbed basic dyes, BY11 and MR, with similar capacities and much better than their commercial counterpart. From previous works (Pereira, et al., 2003, Moreno-Castilla, 2004), for anionic dyes (acid dyes), surface basicity affected the dye adsorption directly. This was caused by the oxygen-free Lewis base sites related to the delocalised π-electrons of the defective grapheme layers. Similarly, for cationic dyes (basic dyes) the carboxylic groups of the carbon surface were more effective. However, the interactions with the π-electrons of the grapheme layers still played a major role. 3.0

2.5

2.0 qe (mmol/g)

Copyright © 2010. Nova Science Publishers, Incorporated. All rights reserved.

Table 2. Physical Properties of the activated carbon produced from bamboo

1.5

1.0

0.5

0.0 0.0

0.2

0.4

0.6

0.8

1.0

1.2

1.4

1.6

1.8

Ce (mmol/L)

HSA1

HSA2

LSA

F400

Figure 3. Plot of qe against Ce for the adsorption of MR onto bamboo produced carbons. Water Production and Wastewater Treatment, Nova Science Publishers, Incorporated, 2010. ProQuest Ebook Central,

Adsorption of Basic Dyes by Activated Carbon from Waste Bamboo

123

2.5

qe (mmol/g)

2.0

1.5

1.0

0.5

0.0 0.00

0.05

0.10

0.15

0.20

0.25

0.30

0.35

0.40

Ce (mmol/L) HSA1

HSA2

LSA

F400

Figure 4. Plot of qe against Ce for the adsorption of BY11 onto bamboo produced carbons.

Copyright © 2010. Nova Science Publishers, Incorporated. All rights reserved.

Equilibrium Adsorption Isotherm Modelling In order to optimise the design of a sorption system to remove the dyes, it is important to establish the most appropriate correlation for the equilibrium curves. The experimental data of the amount of sorbate adsorbed on the sorbent are substituted into an equilibrium isotherm model to determine the best-fit model for the sorption system. Using this relationship, any variation in the concentration of dye on the adsorbent with the concentration of dye in solution is correlated. Three widely used models are Langmuir (Langmuir, 1918), Freundlich (Freundlich, 1906) and Redlich-Peterson (Redlich and Peterson, 1959). Langmuir Isotherm: Langmuir (Langmuir, 1918) proposed a theory to describe the adsorption of gas molecules onto metal surfaces. Langmuir’s model of adsorption depends on the assumption that intermolecular forces decrease rapidly with distance and consequently predicts the existence of monolayer coverage of the adsorbate at the outer surface of the adsorbent. The saturated or monolayer (as Ct → ∞) capacity can be represented by the expression: qe =

K LCe 1 + aLCe

(2)

where qe is solid phase sorbate concentration at equilibrium (mmol/g), Ce is aqueous phase sorbate concentration at equilibrium (mmol/L), KL is Langmuir isotherm constant (L/g), aL is Langmuir isotherm constant (L/mmol) and KL/aL gives the theoretical monolayer saturation capacity, Q0. The Langmuir equation is applicable to homogeneous sorption where the sorption of each sorbate molecule onto the surface has equal sorption activation energy. The Langmuir equation obeys Henry’s Law at low concentration; when the concentration is very low, aLCe is far smaller than unity, it implies qe = KLCe, hence, it is analogous to Henry’s Law. Therefore, a linear expression of the Langmuir equation is:

Water Production and Wastewater Treatment, Nova Science Publishers, Incorporated, 2010. ProQuest Ebook Central,

L. S. Chan, W. H. Cheung, S. J. Allen et al.

124

Ce a 1 = + L Ce qe KL KL

(3)

Therefore, a plot of Ce/qe versus Ce gives a straight line of slope aL/KL and intercept 1/KL, where KL/aL gives the theoretical monolayer saturation capacity, Q0. Freundlich Isotherm: The Freundlich (Freundlich, 1906) equation is an empirical equation employed to describe heterogeneous systems, in which it is characterised by the heterogeneity factor 1/n. When n=1/n, the Freundlich equation reduces to Henry’s Law. Hence, the empirical equation can be written: 1

(4)

qe = K F Cen

where qe is solid phase sorbate concentration in equilibrium (mmol/g), Ce is liquid phase sorbate concentration in equilibrium (mmol/L), KF is Freundlich constant (L/mg1-1/n/g) and 1/n is the heterogeneity factor. This isotherm is another form of the Langmuir approach for adsorption on an “amorphous” surface. The amount adsorbed material is the summation of adsorption on all sites. The Freundlich isotherm is derived by assuming an exponential decay energy distribution function inserted in to the Langmuir equation. It describes reversible adsorption and is not restricted to the formation of the monolayer. A linear form of the Freundlich expression can be obtained by taking logarithms of Equation (4),

Copyright © 2010. Nova Science Publishers, Incorporated. All rights reserved.

ln q e = ln K F +

1 ln C e n

(5)

Therefore, a plot of ln qe versus ln Ce enables the constant KF and exponent 1/n to be determined. Redlich-Peterson Isotherm: Redlich and Peterson (Redlich and Peterson, 1959) incorporate three parameters into an empirical isotherm. The Redlich-Peterson isotherm model combines elements from both the Langmuir and Freundlich equation and the mechanism of adsorption is a hybrid one and does not follow ideal monolayer adsorption. The Redlich-Peterson equation is widely used as a compromise between Langmuir and Freundlich systems.

qe =

K R Ce 1 + aR Ceβ

(6)

where qe is solid phase sorbate concentration in equilibrium (mmol/g), Ce is liquid phase sorbate concentration in equilibrium (mmol/L), KR is Redlich-Peterson isotherm constant (L/g), aR is Redlich-Peterson isotherm constant (L/mg1-1/β) and β is the exponent which lies between 1 and 0.

Water Production and Wastewater Treatment, Nova Science Publishers, Incorporated, 2010. ProQuest Ebook Central,

Adsorption of Basic Dyes by Activated Carbon from Waste Bamboo

125

Linear Approach

Copyright © 2010. Nova Science Publishers, Incorporated. All rights reserved.

Langmuir Isotherm Langmuir’s model of adsorption depends on the assumption that intermolecular forces decrease rapidly with distance and consequently predicts the existence of monolayer coverage of the adsorbate at the outer surface of the adsorbent. The adsorption data were analysed according to the linear form of the Langmuir isotherm equation. (Equation 3). The plots of the specific sorption (Ce/qe) against the equilibrium concentration for the four dye-carbons system are shown in Figures 5 to 6, respectively. All the isotherms were found to be linear over the whole range of the concentration studied and their respective isotherm constants and the correlation coefficients, R2, are shown in the Table 3. The extremely high R2 values reflect that the adsorption data follows the Langmuir model very closely. The Langmuir monolayer capacity Qo represents the saturation capacity of acid dyes in each of the systems and the values for the adsorption system studied are also shown in Table 3. Freundlich Isotherm The Freundlich equation predicts that the dye concentrations on the adsorbent will increase so long as there is an increase in the dye concentration in the liquid. However, the experimental data in the present systems indicate that there is a limiting value of the solid phase concentration. By plotting ln qe versus ln Ce yields the following graphs from the linear transformation of the Freundlich equation. Figures 7 to 8 show the logarithmic plots of the Freundlich expression for the selected dye-carbon adsorption system. The figures exhibit deviation from linearity on the Freundlich linear plot for the whole concentration range. However, for the system of MR adsorbed onto HSA1, the linear Freundlich plots can be divided into regions, i.e. region 1 and region 2, and this interpretation results in better fits to the experimental data at the higher concentration region 2 as shown in Figure 9. The two may be characteristic of two different surfaces and indicative of a wide variation in the two binding site energies. In the initial stage of adsorption the dyes will find their strongest orientation for strong binding on the extensively available carbon surface. Table 3. Langmuir sorption isotherm constant for different dye systems Dye

Carbon

KL (L/g)

aL(L/mmol)

Qo (mmol/g)

R2

MR

HSA1

6.667E+02

3.525E+02

1.891

0.999

HSA2

9.091E+01

5.308E+01

1.713

0.996

LSA

9.709E+01

2.756E+02

0.352

0.998

F400

-1.250E+03

-2.271E+03

0.551

0.990

HSA1

1.667E+03

7.952E+02

2.099

0.998

HSA2

5.263E+02

2.850E+02

1.848

0.994

LSA

6.711E+01

1.538E+02

0.436

0.992

F400

2.128E+02

4.497E+02

0.473

0.999

BY11

Water Production and Wastewater Treatment, Nova Science Publishers, Incorporated, 2010. ProQuest Ebook Central,

L. S. Chan, W. H. Cheung, S. J. Allen et al.

126 5.0 4.5 4.0

Ce/qe (g/L)

3.5 3.0 2.5 2.0 1.5 1.0 0.5 0.0 0.0

0.2

0.4

0.6

0.8

1.0

1.2

1.4

1.6

1.8

Ce (mmol/L) HSA1

HSA2

LSA

F400

Figure 5. Langmuir Isotherm Linear Plots for MR single component system.

0.8 0.7

Ce/qe (g/L)

0.6 0.5 0.4 0.3 0.2

0.0 0.00

0.05

0.10

0.15

0.20

0.25

0.30

0.35

Ce (mmol/L) HSA1

HSA2

LSA

F400

Figure 6. Langmuir Isotherm Linear Plots for BY11 single component system.

2.0 1.5 1.0 0.5 0.0 In qe

Copyright © 2010. Nova Science Publishers, Incorporated. All rights reserved.

0.1

-16.0

-14.0

-12.0

-10.0

-8.0

-6.0

-4.0

-2.0

-0.5 -1.0 -1.5 -2.0 -2.5 -3.0

In Ce HSA1

HSA2

LSA

F400

Figure 7. Freundlich Isotherm Linear Plots for MR single component system. Water Production and Wastewater Treatment, Nova Science Publishers, Incorporated, 2010. ProQuest Ebook Central,

0.0

Adsorption of Basic Dyes by Activated Carbon from Waste Bamboo 1.0

0.5

0.0 -10.0

-8.0

-6.0

-4.0

-2.0

0.0

In qe

-12.0

-0.5

-1.0

-1.5

-2.0 In Ce

HSA1

HSA2

LSA

F400

Figure 8. Freundlich Isotherm Linear Plots for BY11 single component system.

1.5 1.0 y = 0.0869x + 0.6994 R2 = 0.9852

0.5 0.0

-10

-8

-6

In qe

-12

-4 y = 0.1834x + 0.9334 R2 = 0.6988

-2

0 -0.5 -1.0 -1.5

y = 1.3081x + 12.245 R2 = 0.5377

-2.0 -2.5

Copyright © 2010. Nova Science Publishers, Incorporated. All rights reserved.

-3.0 In Ce Expermential Value

Linear (Expermential Value)

Linear (stage 1)

Linear (stage 2)

Figure 9. Freundlich multi-stage model for MR and HSA1 adsorption system.

Table 4. Freundlich sorption isotherm constants for different dye systems Dye

Carbon

bF

KF (L/mg1-1/n/g)

R2

MR

HSA1

1.834E-01

2.543E+00

0.699

HSA2

8.150E-02

1.654E+00

0.965

LSA

-8.000E-03

3.497E-01

0.081

F400

1.402E-01

7.267E-01

0.845

HSA1

2.413E+00

1.130E-01

0.998

HSA2

2.012E+00

1.232E-01

0.961

LSA

5.215E-01

1.454E-01

0.934

F400

6.335E-01

1.543E-01

0.986

BY11

Water Production and Wastewater Treatment, Nova Science Publishers, Incorporated, 2010. ProQuest Ebook Central,

127

L. S. Chan, W. H. Cheung, S. J. Allen et al.

128

Table 5. Multi region Freundlich sorption isotherm constants for dye systems System MR + HSA1

Region (1) (2) Whole

bF 1.308E+00 8.690E-02 2.543E+00

KF (L/mg1-1/n/g) 2.079E+05 2.013E+00 1.834E-01

R2 0.538 0.985 0.699

Eventually, the approach of the dye molecules from the bulk solution will become more restrictive due to the steric hindrance of adsorbed dye molecules already occupying much of the surface sites. It is possible that now only an “end-on” approach by the dye molecule might be feasible and only a weaker adsorption bond is possible. Table 4 shows the Freundlich sorption isotherm constants, bF and KF, and the correlation coefficients, R2 for all the dyecarbon adsorption systems while Table 5 shows the predicted isotherm parameters for the Freundlich multi-stage models for the MR with HSA1 systems.

Non-Linear Approach Due to the inherent bias resulting from the linear transformation of the two parameter equilibrium isotherms, it was decided that a nonlinear approach of the two and three parameter isotherm models by minimising the sum of squares of the error (SSE) function. SSE = ∑ (qe ,cal − qe ,exp )i n

2

(7)

Copyright © 2010. Nova Science Publishers, Incorporated. All rights reserved.

i =1

where qe,calc are the theoretical adsorbed solid phase concentrations of sorbate on sorbent, which have been calculated from one of the isotherm equations and qe,meas are the experimentally determined adsorbed sorbate concentrations obtained from equation (1) using the experimentally measured equilibrium sorbate liquid phase concentrations, Ce. A trial and error procedure was used to determine the isotherm parameters by minimising the error values through the application of the Solver add-in from the spreadsheet software, Microsoft Excel. Although this is the most common error function in use, it has one major drawback. Isotherm parameters derived using this error function will provide a better fit as the magnitude of the errors and thus the squares of the errors increase – biasing the fit towards the data obtained at the high end of the concentration range. The experimental data were fitted into Langmuir, Freundlich, Redlich-Peterson equations to determine which isotherm gives the best correlation to experimental data. Tables 6 and 7 show the values of the parameters for the three isotherm equations. By comparing the SSE of three isotherm equations for BY11, the Langmuir equation fits reasonably well for all carbons. The Freundlich equation provides the best fit of the bamboo carbons. This indicates the possibility of reversible adsorption and is not restricted to the formation of the monolayer. It predicts that the dye concentrations on the adsorbent will increase so long as there is an increase in the dye concentration in the liquid. However, the Redlich-Peterson equation provides the best correlation for the F400.

Water Production and Wastewater Treatment, Nova Science Publishers, Incorporated, 2010. ProQuest Ebook Central,

Adsorption of Basic Dyes by Activated Carbon from Waste Bamboo

129

Table 6. Equilibrium isotherm models’ parameters for BY11- Carbon system

Langmuir Isotherm KL (L/g) aL (L/mmol) Qo (mmol/g) SSE Freundlich Isotherm KF (L/mg1-1/n/g) 1/n SSE Redlich-Peterson Isotherm KR (L/g) aR (L/mg1-1/β) β SSE

HSA1

HSA2

LSA

F400

3899.367 2120.589 1.839 1.931

1191.034 791.663 1.504 1.755

24.546 54.027 0.454 0.075

259.715 567.896 0.457 0.020

2.423 0.114 1.292

2.086 0.134 1.501

0.527 0.151 0.074

0.620 0.146 0.019

5390.515 2366.185 0.920 1.731

4585.060 2248.169 0.880 1.522

56.419 111.499 0.899 0.074

570.0003 1013.323 0.903 0.018

This suggests a reasonable fixed value for the sorption activation energy, which could correspond to the chelation bond energy between the dye ion and surface of the carbon, most likely with a lone pair of electrons on the carbon surface. For MR – carbon systems, similar to the BY11 system, the Langmuir equation provides reasonable correlation for all carbons. The Redlich-Peterson equation provides the best fit for all carbons except HSA2 in which Freundlich is the best fit model.

Copyright © 2010. Nova Science Publishers, Incorporated. All rights reserved.

Table 7. Equilibrium isotherm models’ parameters for MR - Carbon system

Langmuir Isotherm KL (L/g) aL (L/mmol) Qo (mmol/g) SSE Freundlich Isotherm KF (L/mg1-1/n/g) 1/n SSE Redlich-Peterson Isotherm KR (L/g) aR (L/mg1-1/β) β SSE

HSA1

HSA2

LSA

F400

11852.972 6994.135 1.695 0.514

6206.817 3958.791 1.568 0.878

18091.685 51278.851 0.353 0.050

1279.597 2338.480 0.547 0.073

2.144 0.132 0.794

1.661 0.085 0.701

0.353 0.000 0.050

0.684 0.105 0.063

17813.518 9199.995 0.940 0.166

17331.201 10464.509 0.927 0.714

9141.207 25907.075 1.000 0.050

7952.733 13226.475 0.945 0.054

Water Production and Wastewater Treatment, Nova Science Publishers, Incorporated, 2010. ProQuest Ebook Central,

130

L. S. Chan, W. H. Cheung, S. J. Allen et al.

CONCLUSION The preparation and investigation of high BET surface area activated carbons from scrap construction bamboo by low temperature chemical activation has been demonstrated. The high surface area carbon shows nearly three times higher adsorption capacities for both dyes than the commercial carbon, F400. Due to its smaller molecule size, both high surface area carbons have higher adsorption capacities for BY11. However, the low surface area carbon shows poor adsorption for both dyes. The surface area of the carbon plays an important role in the adsorption of the dyes while the porosity has little effect. For BY11, the Freundlich equation provides the best fit for the bamboo carbon while the Redlich-Peterson is the best model to describe the F400 system. For MR, the Redlich-Peterson provides the best isotherm equation for all the carbons except for HSA2 in which the Freundlich equation provides the best correlation. For both basic dyes, the adsorption mechanisms are heterogeneous in nature. It is possible more than one mechanism is involved in the adsorption process.

ACKNOWLEDGMENT The authors would like to acknowledge the support of the Research Grant Council of Hong Kong SAR, the Innovation and Technology Fund of Hong Kong SAR, Hong Kong University of Science and Technology and Green Island International Ltd.

Copyright © 2010. Nova Science Publishers, Incorporated. All rights reserved.

REFERENCES Abe, I.; Fukuhara, T.; Iwasaki, S.; Yasuda, K.; Nakagawa, K.; Iwata, Y.; Kominami, H.; Kera, Y., Development of a high density carbonaceous adsorbent from compressed wood. Carbon 2001, 39 (10), 1485-1490. Ahmedna, M.; Marshall, W. E.; Rao, R. M., Production of granular activated carbons from select agricultural by-products and evaluation of their physical, chemical and adsorption properties. Bioresour. Technol. 2000, 71 (2), 113-123. Al-Degs, Y.; Khraisheh, M. A. M.; Allen, S. J.; Ahmad, M. N., Effect of carbon surface chemistry on the removal of reactive dyes from textile effluent. Water Res. 2000, 34 (3), 927-935. Annadurai, G.; Lee, D. J.; Juang, R. S., Box-Behnken studies on dye removal from water using chitosan and activated carbon adsorbents. J. Chin. Inst. Chem. Eng. 2000, 31 (6), 609-615. Asada, T.; Ishihara, S.; Yamane, T.; Toba, A.; Yamada, A.; Oikawa, K., Science of bamboo charcoal: Study on carbonizing temperature of bamboo charcoal and removal capability of harmful gases. J. Health Sci. 2002, 48 (6), 473-479. Bansal, R. P.; Donnet, J.; Stoeckli, F., Active Carbon. Marcel Dekker: New York, 1988. Brunauer, S.; Emmett, P. H.; Teller, E., Adsorption of gases in multimolecular layers. J. Am. Chem. Soc. 1938, 60, 309-319. Chern, J. M.; Wu, C. Y., Adsorption of binary dye solution onto activated carbon: Isotherm and breakthrough curves. J. Chin. Inst. Chem. Eng. 1999, 30 (6), 507-514.

Water Production and Wastewater Treatment, Nova Science Publishers, Incorporated, 2010. ProQuest Ebook Central,

Copyright © 2010. Nova Science Publishers, Incorporated. All rights reserved.

Adsorption of Basic Dyes by Activated Carbon from Waste Bamboo

131

Corapcioglu, M. O.; Huang, C. P., The Adsorption of Heavy-Metals onto Hydrous Activated Carbon. Water Res. 1987, 21 (9), 1031-1044. Crini, G., Non-conventional low-cost adsorbents for dye removal: A review. Bioresour. Technol. 2006, 97 (9), 1061-1085. Freundlich, H., Concerning adsorption in solutions. Z. Physik. Chem. A 1906, 57 (4), 385470. Garg, V. K.; Gupta, R.; Yadav, A. B.; Kumar, R., Dye removal from aqueous solution by adsorption on treated sawdust. Bioresour. Technol. 2003, 89 (2), 121-124. Horvath, G.; Kawazoe, K., Method for the Calculation of Effective Pore-Size Distribution in Molecular-Sieve Carbon. J. Chem. Eng. Jpn. 1983, 16 (6), 470-475. Jagtoyen, M.; Derbyshire, F., Some Considerations of the Origins of Porosity in Carbons from Chemically Activated Wood. Carbon 1993, 31 (7), 1185-1192. Jagtoyen, M.; Derbyshire, F., Activated carbons from yellow poplar and white oak by H3PO4 activation. Carbon 1998, 36 (7-8), 1085-1097. Juang, R. S.; Tseng, R. L.; Wu, F. C.; Lin, S. J., Use of chitin and chitosan in lobster shell wastes for color removal from aqueous solutions. J. Environ. Sci. Health Part A-Environ. Sci. Eng. Toxic Hazard. Subst. Control 1996, 31 (2), 325-338. Kannan, N.; Sundaram, M. M., Kinetics and mechanism of removal of methylene blue by adsorption on various carbons - a comparative study. Dyes Pigment. 2001, 51 (1), 25-40. Laine, J.; Calafat, A.; Labady, M., Preparation and Characterization of Activated Carbons from Coconut Shell Impregnated with Phosphoric-Acid. Carbon 1989, 27 (2), 191-195. Langmuir, I., The adsorption of gases on plane surfaces of glass, mica and platinum. J. Am. Chem. Soc. 1918, 40 (9), 1361 - 1403. Marc, R., Asian texile dye makers are a growing power in changing market. CandEN 1996, 73, 10-12. Marshall, W. E.; Ahmedna, M.; Rao, R. M.; Johns, M. M., Granular activated carbons from sugarcane bagasse: production and uses. Int. Sugar J. 2000, 102 (1215), 147-151. Moreno-Castilla, C., Adsorption of organic molecules from aqueous solutions on carbon materials. Carbon 2004, 42, 83-93. Namasivayam, C.; Radhika, R.; Suba, S., Uptake of dyes by a promising locally available agricultural solid waste: coir pith. Waste Manage. 2001, 21 (4), 381-387. Ohe, K.; Nagae, Y.; Nakamura, S.; Baba, Y., Removal of nitrate anion by carbonaceous materials prepared from bamboo and coconut shell. J. Chem. Eng. Jpn. 2003, 36 (4), 511515. Pereira, M. F. R.; Soares, S. F.; Orfao, J. J. M.; Figueiredo, J. L., Adsorption of dyes on activated carbons: influence of surface chemical groups. Carbon 2003, 41 (4), 811-821. Ramesh, A.; Lee, D. J.; Wong, J. W. C., Adsorption equilibrium of heavy metals and dyes from wastewater with low-cost adsorbents: A review. J. Chin. Inst. Chem. Eng. 2005, 36 (3), 203-222. Redlich, O.; Peterson, D. L., A Useful Adsorption Isotherm. J. Phys. Chem. 1959, 63 (6), 1024. Tsai, W. T.; Chang, C. Y.; Lee, S. L., A low cost adsorbent from agricultural waste corn cob by zinc chloride activation. Bioresour. Technol. 1998, 64 (3), 211-217. Valix, M.; Cheung, W. H.; McKay, G., Preparation of activated carbon using low temperature carbonisation and physical activation of high ash raw bagasse for acid dye adsorption. Chemosphere 2004, 56 (5), 493-501.

Water Production and Wastewater Treatment, Nova Science Publishers, Incorporated, 2010. ProQuest Ebook Central,

132

L. S. Chan, W. H. Cheung, S. J. Allen et al.

Copyright © 2010. Nova Science Publishers, Incorporated. All rights reserved.

Walker, G. M.; Hansen, L.; Hanna, J. A.; Allen, S. J., Kinetics of a reactive dye adsorption onto dolomitic sorbents. Water Res. 2003, 37 (9), 2081-2089. Wu, F. C.; Tseng, R. L.; Juang, R. S., Preparation of activated carbons from bamboo and their adsorption abilities for dyes and phenol. J. Environ. Sci. Health Part A-Toxic/Hazard. Subst. Environ. Eng. 1999, 34 (9), 1753-1775.

Water Production and Wastewater Treatment, Nova Science Publishers, Incorporated, 2010. ProQuest Ebook Central,

In: Water Production and Wastewater Treatment Editors: B. Antizar-Ladislao and R. Sheikholeslami

ISBN 978-1-61728-503-5 © 2011 Nova Science Publishers, Inc.

Chapter 8

INVESTIGATIONS ON ARSENIC ADSORPTION ONTO DOLOMITIC SORBENTS Y. Salameh1, M.N.M. Ahmad, S.J. Allen, and G.M. Walker School of Chemistry and Chemical Engineering, Queen’s University Belfast, Belfast BT9 5AG, Northern Ireland, UK

Copyright © 2010. Nova Science Publishers, Incorporated. All rights reserved.

ABSTRACT Arsenic is present in potable water in many areas in the world as a result of both natural impacts and anthropogenic discharge, with arsenic bearing waters requiring proper treatment before use. At present, there is a considerable interest in studying new sorbent materials for the removal of arsenic from aqueous solutions. This work discusses the feasibility of arsenic uptake onto raw dolomite which is considered to be a potential inexpensive adsorbent. Experimental investigations were undertaken in equilibrium isotherm and kinetic systems in order to evaluate the adsorption capacity by taking into consideration the experimental parameters such as: pH; initial solute concentration; massvolume ratio; particle size of adsorbent; contact time, the effect of various ions present and the effect of changing the temperature. The equilibrium time was determined to be 5 days for dolomite. Desorption studies were also undertaken. The data for the dolomite-As were compared with granular activated carbon in an identical set of experiments. The data were mathematically described using empirical equilibrium isotherm models, namely Langmuir and Freundlich models. The maximum arsenic removal with dolomite was found at pH 2 and was dependent on the dosage of dolomite, adsorbent particle size and the presence of various anions. For the kinetic Experiments the data were mathematically described using adsorption kinetic models, namely pseudo first-order and pseudo second-order models. Thermodynamic results indicate that the adsorption follows an exothermic chemisorption process. The experimental data indicate successful removal of As(V) ion from aqueous solution indicating that dolomite may be an alternative low cost adsorbent for As(V).

1Tel. +44 (0) 2890 974253, Fax. +44 (0)2890 974627 e-mail [email protected]. Water Production and Wastewater Treatment, Nova Science Publishers, Incorporated, 2010. ProQuest Ebook Central,

134

Y. Salameh, M. N. M. Ahmad, S. J. Allen et al.

1. INTRODUCTION

Copyright © 2010. Nova Science Publishers, Incorporated. All rights reserved.

1.1. Arsenic in Groundwater Arsenic bearing compounds are toxic to all living organisms. The presence of arsenic is a worldwide environmental problem with regard to drinking water (Guan et al., 2008). Significant problems occur when arsenic is widespread but undetected because its long-term health effects are caused by chronic as opposed to acute exposure. Arsenic has been linked to cancer of the bladder, lungs, skin, kidney, nasal passages, liver, and prostate (Roy, 2008; Biswas et al., 2008). Non-cancer effects include gangrene, limb loss, cardiovascular and pulmonary disease, endocrine and haematological disorders, and reproductive/developmental problems. In addition, arsenic is considered an accumulative enabler, meaning people who are predisposed to various cancers, diabetes, high blood pressure and other ailments are more likely to contract these illnesses (Nguyen et al., 2007). Arsenic occurs in various mineral forms, of which 60% are arsenates, 20% are sulphides and sulphosalts, 10% are oxides and the remainder are arsenides, native elements and metal alloys (Mkandawire and Dudel, 2005). Several toxicological studies have demonstrated that the toxicity of arsenic is dependent on its chemical forms, oxidation state, physical state (gas or solution), rate of absorption into cells, rate of elimination and its chemical nature. Arsenic exists in several states of oxidation: As(0) or as ion forms like As(V) arsenate, As(III) arsenite and As(III) arsine. It is generally recognized that the soluble inorganic arsenicals are more toxic than the organic ones, and the inorganic As(III) species are more toxic than the inorganic As(V) (Sanchez de la Campa et al., 2008). At moderate or high redox potentials, arsenic can be stabilized as a series of pentavalent (arsenate) oxyanions, H3AsO4, H2AsO4−, HAsO4 2− and AsO43−. Thus, to eliminate these toxic metals from water, a potential adsorbent must have the anion exchange properties (Yusof and Malek, 2008). Increased concentrations of arsenic in natural waters have been reported in many areas of the world such as, in South East Asia (Bangladesh, Vietnam, West Bengal-India, Nepal, Cambodia, Mongolia, China, Thailand, Pakistan and Taiwan), in Central and South America (Mexico, Chile and Argentina) and in North America (USA and Canada) and in Australia. Elevated concentrations of arsenic have been also found in various European countries, i.e., Finland, Hungary, Germany, Croatia, Romania, Italy, Spain and Greece (Biterna et al., 2007). Arsenic is introduced into water by natural processes such as weathering reactions, biological activity and volcanic emission; it also enters the water cycle through discharges of various industries such as smelting, petroleum refinery, fertilizers, insecticides, herbicides, glass and ceramic manufacturing industries (Shao et al., 2008). Arsenic is mainly ingested through contaminated water supplies where it generally occurs as As(V) and As(III), depending on pH and redox conditions. Due to the high toxicity and the widespread occurrence of arsenic, it is strictly controlled by environmental regulations in most areas of the world. The World Health Organization has defined a restrictive limit of 10 ppb for drinking water (WHO, 2006). To help achieve this consent limit, various treatment technologies to remove arsenic from drinking water have been designed such as: coagulation; ion exchange; reverse osmosis; liquid-liquid extraction; ultra-filtration; and adsorption (Guan et al., 2008; Biswas et al., 2008). However, in many

Water Production and Wastewater Treatment, Nova Science Publishers, Incorporated, 2010. ProQuest Ebook Central,

Investigations on Arsenic Adsorption onto Dolomitic Sorbents

135

areas of the world there is still in a necessity for appropriate technology, which is inexpensive, simple to use and easily applied.

1.2. The Dolomite Group of Minerals The Dolomite is considered a promising low cost adsorbent. It is composed of minerals with an unusual trigonal bar 3 symmetry. The general formula of this group is AB(CO3)2, where A can be either calcium, barium and/or strontium and the B can be either iron, magnesium, zinc and/or manganese. The structure of the Dolomite Group is taken from the Calcite Group structure. The Calcite Group structure is layered with alternating carbonate layers and metal ion layers. The structure of the Dolomite Group minerals is layered in such a way that the A metal ions occupy one layer which is followed by a carbonate layer which is followed by the B metal ion layer followed by another carbonate (CO3) layer, etc. The layering is of the form: |A|CO3|B|CO3|A|CO3|B|CO3|A|... This ordered layering of different or non-equivalent ions causes a loss of the two fold rotational axes and mirror planes that are present in the Calcite Group structure (Walker et al., 2004; Walker et al., 2005). The amount of calcium and magnesium in most specimens is equal, but occasionally one element may have a slightly greater presence than the other. Small amounts of iron and manganese are sometimes also present (Duffy et al., 2005).

Copyright © 2010. Nova Science Publishers, Incorporated. All rights reserved.

1.3. Analysis of Adsorption Equilibrium Isotherms Two of the most commonly used isotherm models have been used in this work, namely, the Langmuir and Freundlich models. The form of the Langmuir equation can be represented as follows:

qe =

Qmax K L Ce 1 + K L Ce

(1)

Or in the linear from:

Ce 1 Ce = + qe K L Qmax Qmax

(2)

where: Ce is the equilibrium concentration of As(V) in solution (mg/L); qe is the solid phase solute concentration (mg g-1); Qmax is the maximum arsenic uptake capacity (mg g-1); KL is the equilibrium constant. The Freundlich model has the following form

qe = k F Ce

n

Water Production and Wastewater Treatment, Nova Science Publishers, Incorporated, 2010. ProQuest Ebook Central,

(3)

Y. Salameh, M. N. M. Ahmad, S. J. Allen et al.

136

Or in the linear from:

log qe = log k F + n log Ce

(4)

where: KF is the constant indicative of the relative adsorption capacity of the adsorbent (mg g1 ); n is the constant indicative of the intensity of the adsorption and it is dimensionless. Percentage solute removal was then calculated as follows: Removal(100%) =

Co − Ce * 100% Co

(5)

where: Co and Ce are the initial and equilibrium solute concentrations respectively.

1.4. Kinetic Modelling The batch experimental data were applied to selected adsorption kinetic models, namely pseudo first-order and pseudo second-order models. The pseudo first-order (Lagergren firstorder) rate equation is as follows:

Copyright © 2010. Nova Science Publishers, Incorporated. All rights reserved.

ln(qe − qt) = ln qe − K1ads t

(6)

where: qe and qt are the amount of adsorbate adsorbed (μg/g) at equilibrium and at time t (min), K1ads is the adsorption rate constant. The values of K1ads and qe were calculated from the intercept and slope of the plots of ln(qe - qt) versus t. The pseudo second-order equation is also based on the sorption capacity of the solid phase and is expressed as;

dq = K2ad(qe − qt)2 dt

(7)

where K2ad is the rate constant of second-order adsorption. For the same boundary conditions the integrated form becomes:

t 1 1 + ( )t = 2 qt qe K 2 ad q e

(8)

If second-order kinetics is applicable, the plot of t/q against t of should give a linear relationship, from which qe and K2ad can be determined from the slope and intercept of plot.

Water Production and Wastewater Treatment, Nova Science Publishers, Incorporated, 2010. ProQuest Ebook Central,

Investigations on Arsenic Adsorption onto Dolomitic Sorbents

137

1.5. Thermodynamic Modelling To calculate the thermodynamic activation parameters such as enthalpy of activation, ∆ H◦, entropy of activation, ∆S◦, and free energy of activation, ∆G◦, the Eyring equation was applied (Al-Ghouti et al., 2005),

⎛ K 2 ads ⎞ ⎡ ⎛⎜ k B ⎟ = ⎢ln ⎝ T ⎠ ⎢⎣ ⎜⎝ h p

ln ⎜

⎞ ΔS o ⎤ ΔH o ⎟+ ⎥− ⎟ R ⎥⎦ R ⎠

⎛1⎞ ⎜ ⎟, ⎝T ⎠

(9)

where: ∆ G◦ = ∆ H◦ - T∆ S◦ (10) where kB is the Boltzmann constant (1.3807 × 10−23 J/K), hP is the Planck constant (6.6261 × 10−34 J s), R is the ideal gas constant (8.314 J g-1 K-1), K2ad is the pseudo-second-order constant (k2) for arsenic adsorption. The activation energy of arsenic adsorption onto the dolomite can be calculated by the relationship : ln(K2ad) = ln(Ko)−

E⎛1⎞ ⎜ ⎟ R ⎝T ⎠

(11)

Copyright © 2010. Nova Science Publishers, Incorporated. All rights reserved.

where Ko is the rate constant of adsorption (g/mg min). Plotting the pseudo-second-order constant against the reciprocal temperature gives a reasonably straight line, the gradient of which is −E/R.

2. EXPERIMENTAL MATERIALS AND METHODS 2.1. Adsorbent Characterisation The dolomite used in this study was mined from a deposit in County Fermanagh, Northern Ireland. The typical chemical composition of the dolomite in the deposit was 44% MgCO3 and 53% CaCO3. Table 1. General properties of activated carbon used in this study specification BET surface area, m2/g

value 941

External surface area, m2/g

396

Micropore surface area, m2/g

541

Water Production and Wastewater Treatment, Nova Science Publishers, Incorporated, 2010. ProQuest Ebook Central,

138

Y. Salameh, M. N. M. Ahmad, S. J. Allen et al.

Figure 1. SEM of surface of raw dolomite sample at a magnitude of 2000mm×25 mm.

The dolomite was ground and sieved on a series of test sieves. A sample of dolomite was analysed for specific surface area using BET nitrogen adsorption employing a Nova 4200e, surface area and pore size analyser (Quantachrome Instruments), the surface area was found to be 17.36(m2g-1) for particle size 1.0-1.2 mm. Figure 1 shows a SEM image for a raw dolomite sample. It can be observed that the dolomite structure consists of a crystalline structure with inter-special voids. Activated carbon used in the studies was commercially available having specification as given in Table 1.

Copyright © 2010. Nova Science Publishers, Incorporated. All rights reserved.

2.2. Adsorbate Characterisation A 2000ppb stock arsenic solution was prepared in a 1L volumetric flask using deionized water. The stock solution was made from sodium arsenic dibasic heptahydrate (Na2HAsO4.7H2O), Aldrich Chemical Co. Inc, USA. In addition 0.1g NaHCO3 was added to buffer pH fluctuations during the experiment. Most groundwaters have some alkalinity present, which tends to be in the order of 250-600 mg/L HCO3. The addition of bicarbonate is therefore not seen to comprise the experiment. A simple picture of arsenic chemical speciation calculated based on their stability constants has been outlined in Figure 2 (Shao et al., 2008). The monovalent anionic species of arsenate were dominant at pH ranging from 2 to 6, while in the case of arsenite were significant at alkaline region of pH 9–12.

Figure 2. Distribution of arsenate and arsenite as a function of pH (after Shao et al., 2008). Water Production and Wastewater Treatment, Nova Science Publishers, Incorporated, 2010. ProQuest Ebook Central,

Investigations on Arsenic Adsorption onto Dolomitic Sorbents

139

Copyright © 2010. Nova Science Publishers, Incorporated. All rights reserved.

2.3. Experimental Methods Preliminary experiments were conducted to evaluate the impact of using dolomite as an adsorbent for As(III). However, the results indicated very low removal ( 1 indicates that the adsorption bonds are weak, adsorption capacities decrease and unfavourable. The RL values for the range of the arsenic concentrations under investigation have been calculated and found to lie between 0 and 1, which indicated that the sorption of As(V) by raw dolomite is favourable at pH 2 and temperature 22°C. Figures 5 and 6 show the Langmuir and Freundlich isotherms respectively for As(V)activated carbon systems. For Langmuir the value of KL = 3.12 x10-3(L mg-1) and Qmax = 0.625(mg g-1). For Freundlich the values of KF = 1.622(mg g-1) and n = 0.687.

Water Production and Wastewater Treatment, Nova Science Publishers, Incorporated, 2010. ProQuest Ebook Central,

Y. Salameh, M. N. M. Ahmad, S. J. Allen et al.

142

3.5 3

C e /q e [g /l]

2.5 2 1.5 1 0.5 0 0

200

400

600

800

1000

1200

1400

1600

1800

equilibrium concentration of As(V) [ppb]

Figure 6. Langmuir plot for the adsorption of arsenic onto activated carbon. Adsorbent dose: 1g/L; volume of test solution: 50 ml; pH 7; contact time: 5 days.

3 2.5

lo g q e

Copyright © 2010. Nova Science Publishers, Incorporated. All rights reserved.

2 1.5 1 0.5 0 0

0.5

1

1.5

2

2.5

3

3.5

4

log Ce

Figure 7. Freundlich plot for the adsorption of arsenic onto activated carbon. Adsorbent dose: 1g/L; volume of test solution: 50 ml; pH 7; contact time: 5 days.

3.1.3. Effect of Adsorbent Concentration The effect of adsorbent mass to volume of solution on As(V) uptake is illustrated in Figures 8 and 9, which shows that adsorption efficiency of As(V) increased with an increase in mass of raw dolomite.

Water Production and Wastewater Treatment, Nova Science Publishers, Incorporated, 2010. ProQuest Ebook Central,

Investigations on Arsenic Adsorption onto Dolomitic Sorbents

143

90 80 70

removal (%)

60 2000 ppb 50

1000 ppb

40

500 ppb 50 ppb

30 20 10 0 0

0.5

1

1.5

2

2.5

3

3.5

adsorbant concentration (g/L)

Figure 8. The dosage of adsorbent dependence for As(V) adsorption on raw dolomite. Volume of test solution: 50 ml; pH 2; contact time: 5 days.

100 90 80 70 2000 ppb

60

1000 ppb

50

500 ppb

40

50 ppb

Copyright © 2010. Nova Science Publishers, Incorporated. All rights reserved.

30 20 10 0 0

1

2

3

4

adsorbant concentration (g/L)

Figure 9. The dosage of adsorbent dependence for As(V) adsorption on activated carbon. Volume of test solution: 50 ml; pH 7; contact time: 5 days.

The increase in the efficiency of removal can be attributed to increased adsorbent surface area available for mass transfer, i.e., more pores and voids for As(V) chemisorption, but also higher concentrations of MgCO3 and CaCO3 which would increase the probability of formation of arsenic oxide and arsenic carbonate, thus increasing in the extent of precipitation as a mechanism of As(V) removal.

3.1.4. Effect of Adsorbent Particle Size A range of particle sizes have been tested in this study in order to understand the mechanism of As(V) removal by raw dolomite. The particle size has been varied from less than 0.335 mm to (0.710-2.00) mm. Figure 10 indicates that the amount of solute adsorbed increases with the decrease in particle size of the adsorbent.

Water Production and Wastewater Treatment, Nova Science Publishers, Incorporated, 2010. ProQuest Ebook Central,

Y. Salameh, M. N. M. Ahmad, S. J. Allen et al.

144

900 800 700 600 dp